Estimate the diamond ratio for a meta-analytic effect, a measure of heterogeneity

Description

CI_diamond_ratio returns the diamond ratio and CI for a meta-analytic effect, the ratio of the random-effects CI width to the fixed-effects CI width. The diamond ratio is a measure of effect-size heterogeneity.

Usage

CI_diamond_ratio(RE, FE, vi, conf_level = 0.95)

Arguments

Details

Calculation of the CI is based on code provided by Maxwell Cairns (see Cairns et al., 2022). Specifically, this function implements what Cairns et al (2022) called the Sub-Q approach, which provides the best CI coverage in simulations. For comparison, this function also returns the CI produced by the bWT-DL approach (which generally has worse performance).

Value

Returns a list with 3 properties:

• diamond_ratio

• LL - lower limit of the conf_level% CI, Sub-Q approach

• UL - upper limit of the conf_level% CI, Sub-Q approach

• LL_bWT_DL - lower limit of the conf_level% CI, bWT-DL approach

• UL_bWT_DL - upper limit of the conf_level% CI, bWT-DL approach

Source

Cairns, Maxwell, Geoff Cumming, Robert Calin‐Jageman, and Luke A. Prendergast. “The Diamond Ratio: A Visual Indicator of the Extent of Heterogeneity in Meta‐analysis.” British Journal of Mathematical and Statistical Psychology 75, no. 2 (May 2022): 201–19. https://doi.org/10.1111/bmsp.12258.

Examples

mydata <- esci::data_mccabemichael_brain

# Use esci to obtain effect sizes and sample variances, storing only raw_data
mydata <- esci::meta_mdiff_two(
  data = mydata,
  comparison_means = "M Brain",
  comparison_ns = "n Brain",
  comparison_sds = "s Brain",
  reference_means = "M No Brain",
  reference_ns = "n No Brain",
  reference_sds = "s No Brain",
  random_effects = FALSE
)$raw_data

# Conduct fixed effects meta-analysis
FE <- metafor::rma(
  data = mydata,
  yi = effect_size,
  vi = sample_variance,
  method="FE"
)
# Conduct random effect meta-analysis
RE <- metafor::rma(
  data = mydata,
  yi = effect_size,
  vi = sample_variance,
  method="DL"
)

# Get the diamond ratio
res <- esci::CI_diamond_ratio(
  RE = RE,
  FE = FE,
  vi = mydata$sample_variance
)

Estimate standardized mean difference (Cohen's d) for an independent groups contrast

Description

CI_smd_ind_contrast returns the point estimate and confidence interval for a standardized mean difference (smd aka Cohen's d aka Hedges g). A standardized mean difference is a difference in means standardized to a standard deviation: \[d = \frac{ \psi }{s}\]

Usage

CI_smd_ind_contrast(
  means,
  sds,
  ns,
  contrast,
  conf_level = 0.95,
  assume_equal_variance = FALSE,
  correct_bias = TRUE
)

Arguments

Value

Returns a list with these named elements:

• effect_size - the point estimate from the sample

• lower - lower bound of the CI

• upper - upper bound of the CI

• numerator - the numerator for Cohen's d_biased; the mean difference in the contrast

• denominator - the denominator for Cohen's d_biased; if equal variance is assumed this is sd_pooled, otherwise sd_avg

• df - the degrees of freedom used for correction and CI calculation

• se - the standard error of the estimate; warning not totally sure about this yet

• moe - margin of error; 1/2 length of the CI

• d_biased - Cohen's d without correction applied

• properties - a list of properties for the result

  Properties

• effect_size_name - if equal variance assumed d_s, otherwise d_avg

• effect_size_name_html - html representation of d_name

• denominator_name - if equal variance assumed sd_pooled otherwise sd_avg

• denominator_name_html - html representation of denominator name

• bias_corrected - TRUE/FALSE if bias correction was applied

• message - a message explaining denominator and correction status

• message_html - html representation of message

Details

It's a bit complicated

A standardized mean difference turns out to be complicated.

First, it has many names:

• standardized mean difference (smd)

• Cohen's d

• When bias in a sample d has been corrected, also called Hedge's g

Second, the choice of the standardizer requires thought:

• sd_pooled - used when assuming all groups have exact same variance

• sd_avg - does not require assumption of equal variance

• other possibilities, too, but not dealt with in this function

The choice of standardizer is important, so it's noted in the subscript:

• d_s – assumes equal variance, standardized to sd_pooled

• d_avg - does not assume equal variance, standardized to sd_avg

A third complication is the issue of bias: d estimated from a sample has a slight upward bias at smaller sample sizes. With total sample size > 30, this slight bias becomes fairly negligible (kind of like the small upward bias in a sample standard deviation).

This bias can be corrected when equal variance is assumed or when the design of the study is simple (2 groups). For complex designs (>2 groups) without the assumption of equal variance, there is now also an approximate approach to correcting bias from Bonett.

Corrections for bias produce a long-run reduction in average bias. Corrections for bias are approximate.

How are d and its CI calculated?

When equal variance is assumed

When equal variance is assumed, the standardized mean difference is d_s, defined in Kline, p. 196:

where psi is defined in Kline, equation 7.8:

and where sd_pooled is defined in Kline, equation 3.11 \[sd_{pooled} = { \frac{ \sum_{i=1}^{a} (n_i -1) s_i^2 } { \sum_{i=1}^{a} (n_i-1) } }\]

The CI for d_s is derived from lambda-prime transformation from Lecoutre, 2007 with code adapted from Cousineau & Goulet-Pelletier, 2020. Kelley, 2007 explains the general approach for linear contrasts.

This approach to generating the CI is 'exact', meaning coverage should be as desired if all assumptions are met (ha!).

Correction of upward bias can be applied.

When equal variance is not assumed

When equal variance is not assumed, the standardized mean difference is d_avg, defined in Bonett, equation 6:

Where sd_avg is the square root of the average of the group variances, as given in Bonett, explanation of equation 6:

If only 2 groups

• The CI is derived from lambda-prime transformation using df and se from Huynh, 1989 – see especially Delacre et al., 2021

• This is also an 'exact' approach, and correction can be applied

If more than 2 groups

• CI is approximated using the approach from Bonett, 2008

• An approximate correction developed by Bonett is used

References

• Bonett D. G. (2023). statpsych: Statistical Methods for Psychologists. R package version 1.4.0. https://dgbonett.github.io/statpsych

• Bonett, D. G. (2018). R code posted to personal website (now removed). Formally at https://people.ucsc.edu/~dgbonett/psyc204.html

• Bonett, D. G. (2008). Confidence Intervals for Standardized Linear Contrasts of Means. Psychological Methods, 13(2), 99–109. doi:10.1037/1082-989X.13.2.99

• Cousineau & Goulet-Pelletier (2020) A review of five techniques to derive confidence intervals with a special attention to the Cohen’s dp in the between-group design. doi:10.31234/osf.io/s2597

• Delacre M, Lakens D, Ley C, Liu L, Leys C (2021) Why Hedges gs based on the non-pooled standard deviation should be reported with Welch’s t-test. doi:10.31234/osf.io/tu6mp

• Huynh, C.-L. (1989). A unified approach to the estimation of effect size in meta-analysis. NBER Working Paper Series, 58(58), 99–104.

• Kelley, K. (2007). Confidence intervals for standardized effect sizes: Theory, application, and implementation. Journal of Statistical Software, 20(8), 1–24. doi:10.18637/jss.v020.i08

• Lecoutre, B. (2007). Another Look at the Confidence Intervals for the Noncentral T Distribution. Journal of Modern Applied Statistical Methods, 6(1), 107–116. doi:10.22237/jmasm/1177992600

See Also

• estimate_mdiff_ind_contrast for friendly version that also returns raw score effect sizes for this design

Examples

# Example from Kline, 2013
#  Data in Table 3.4
#  Worked out in Chapter 7
#  See p. 202, non-central approach
# With equal variance assumed and no correction, should give:
#   d_s = -0.8528028 [-2.121155, 0.4482578]

res <- esci::CI_smd_ind_contrast(
  means = c(13, 11, 15),
  sds = c(2.738613, 2.236068, 2.000000),
  ns = c(5, 5, 5),
  contrast = contrast <- c(1, 0, -1),
  conf_level = 0.95,
  assume_equal_variance = TRUE,
  correct_bias = FALSE
)


# Example from [statpsych::ci.lc.stdmean.bs()] should give:
# Estimate        SE        LL         UL
# Unweighted standardizer: -1.273964 0.3692800 -2.025039 -0.5774878
# Weighted standardizer:   -1.273964 0.3514511 -1.990095 -0.6124317
# Group 1 standardizer:    -1.273810 0.4849842 -2.343781 -0.4426775


res <- esci::CI_smd_ind_contrast(
  means = c(33.5, 37.9, 38.0, 44.1),
  sds = c(3.84, 3.84, 3.65, 4.98),
  ns = c(10,10,10,10),
  contrast = c(.5, .5, -.5, -.5),
  conf_level = 0.95,
  assume_equal_variance = FALSE,
  correct_bias = TRUE
)

Estimate standardized mean difference (Cohen's d1) for a single group

Description

CI_smd_one STILL NEEDS WORK TO VERIFY APPROACH FOR SE and MoE

Usage

CI_smd_one(mean, sd, n, reference_mean, conf_level = 0.95, correct_bias = TRUE)

Arguments

Value

Returns a list with these named elements:

• effect_size - the point estimate from the sample

• lower - lower bound of the CI

• upper - upper bound of the CI

• numerator - the numerator for Cohen's d_biased; the mean difference in the contrast

• denominator - the denominator for Cohen's d_biased; if equal variance is assumed this is sd_pooled, otherwise sd_avg

• df - the degrees of freedom used for correction and CI calculation

• se - the standard error of the estimate; warning not totally sure about this yet

• moe - margin of error; 1/2 length of the CI

• d_biased - Cohen's d without correction applied

• properties - a list of properties for the result

Properties

• effect_size_name - if equal variance assumed d_s, otherwise d_avg

• effect_size_name_html - html representation of d_name

• denominator_name - if equal variance assumed sd_pooled otherwise sd_avg

• denominator_name_html - html representation of denominator name

• bias_corrected - TRUE/FALSE if bias correction was applied

• message - a message explaining denominator and correction status

• message_html - html representation of message

Examples

# example code
res <- esci::CI_smd_one(24.5, 3.65, 40, 20)

Altruism Happiness - Ch12 - from Brethel-Haurwitz and Marsh (2014)

Description

Happiness may not be important just for the person feeling it; happiness may also promote kind, altruistic behavior. Brethel-Haurwitz and Marsh (2014) examined this idea by collecting data on U.S. states. A Gallup poll in 2010 was used to measure each state's well-being index, a measure of mean happiness for the state's residents on a scale from 0 to 100. Next, a kidney donation database for 1999-2010 was used to figure out each state's rate (number of donations per 1 million people) of non-directed kidney donations-that's giving one kidney to a stranger, an extremely generous and altruistic thing to do!

Usage

data_altruism_happiness

Format

data_altruism_happiness

A data frame with 50 rows and 6 columns:

State

factor - State where data was collected

Abbreviation

factor - State where data was collected

Well_Being_2010

numeric - State where data was collected

Well_Being_2013

numeric - State where data was collected

Kidney_Rate, per million population

numeric - State where data was collected

WB Change 2013-2010

numeric - State where data was collected

Source

doi:10.1177/0956797613516148

Anchor Estimate ma - Ch9 - Many Labs replications of Jacowitz and Kahneman (1995)

Description

To what extent does the wording of a question influence one's judgment? In their classic study, Jacowitz and Kahneman (1995) asked participants to estimate how many babies are born each day in the United States. Participants were given either a low anchor (more than 100 babies/day) or a high anchor (less than 50,000 babies/day). Those who saw the low anchor estimated many fewer births/day than those who saw the high anchor, which suggests that the wording can have a profound influence. The correct answer, as it happens, is ~11,000 births/day in 2014. To investigate the extent that these results are replicable, the Many Labs project repeated this classic study at many different labs around the world. You can find the summary data for 30 of these labs in the Anchor Estimate ma data file

Usage

data_anchor_estimate_ma

Format

data_anchor_estimate_ma

A data frame with 30 rows and 9 columns:

Location

factor

M Low

numeric

s Low

numeric

n Low

integer

M High

numeric

s High

numeric

n High

integer

USAorNot

factor

Country

factor

Source

doi:10.1027/1864-9335/a000178

Basol badnews - Ch07 - from Basol et al. (2020)

Description

Climate change? Vaccines? Fake news and conspiracy theories on these and numerous other issues can be highly damaging, but are thriving in this social media age. Trying to debunk a conspiracy theory by presenting facts and evidence often doesn't work, alas. Psychological inoculation, also similar to prebunking, presents a mild form of misinformation, preferably with explanation, in the hope of building resistance to real-life fake news-a sort of vaccine for fake news. The Bad News game is a spin-off from research on psychological inoculation. Basol et al. (2020) assessed the possible effectiveness of this game as a fake news vaccine. At getbadnews.com you can click 'About' for information, or just start playing the game- it's easy and maybe even fun. You encounter mock Twitter (now X) fake news messages that illustrate common strategies for making fake news memorable or believable. You make choices between messages and decide which ones to 'forward' as you try to spread fake news while building your credibility score and number of 'followers'-rather like real life for a conspiracy theorist wanting to spread the word. Compete with your friends for credibility and number of followers. Basol's online participants first saw 18 fictitious fake news tweets and rated each for reliability (accuracy, believability), and also rated their confidence in that reliability rating. Both ratings were on a 1 to 7 scale. Those in the BadNews group then played the game for 15 minutes, whereas those in the Control group played Tetris. Then all once again gave reliability and confidence ratings for the 18 tweets.

Usage

data_basol_badnews

Format

data_basol_badnews

A data frame with 198 rows and 3 columns:

Diff reliability

numeric

Diff confidence

numeric

Condition

factor

Source

doi:10.5334/joc.91

Bem Psychic - Ch13 - from Bem and Honorton (1994)

Description

Daryl Bem was an experienced mentalist and research psychologist, who, a decade earlier, had been one of several outside experts invited to scrutinize the laboratory and experimental procedures of parapsychology researcher Charles Honorton. Bem not only judged them adequate, but joined the research effort and became a coauthor with Honorton. Bem and Honorton (1994) first reviewed early ganzfeld studies and described how the experimental procedure had been improved to reduce the chance that results could be influenced by various possible biases, or leakages of information from sender to receiver. For example, the randomization procedure was carried out automatically by computer, and all stimuli were presented under computer control. Bem and Honorton then presented data from studies conducted with the improved procedure. Table 13.1 presents basic data from 10 studies reported by Bem and Honorton (1994). Participants each made a single judgment, so in Pilot 1, for example, 22 participants responded, with 8 of them giving a correct response. Three pilot studies helped refine the procedures, then four studies used novice receivers. Study 5 used 20 students of music, drama, or dance as receivers, in response to suggestions that creative people might be more likely to show telepathy. Studies 6 and 7 used receivers who had participated in an earlier study. The proportion of hits expected by chance is .25, and Table 13.1 shows that all but Study 1 found proportions higher than .25.

Usage

data_bem_psychic

Format

data_bem_psychic

A data frame with 10 rows and 5 columns:

Study

factor

Participants

factor

N(Trials)

integer

N(Hits)

integer

Proportion Hits

numeric

Source

doi:10.1037/0033-2909.115.1.25

BodyWellF - Ch12 - Body Satisfaction and Well-being data for females from Figure 11.24 right panel

Description

A subset of data_bodywell_fm, reports only for those participants who identified as female. Data is Subjective Wellbeing and Body Satisfaction.

Usage

data_bodywellf

Format

data_bodywellf

A data frame with 59 rows and 2 columns:

Body Satisfaction

numeric

Well-being

numeric

BodyWellFM - Ch12 - Body Satisfaction and Well-being data from Figure 11.1

Description

Survey data from a convenience sample Dominican University students. Reported are measures of Subjective Wellbeing abd Body Satisfaction.

Usage

data_bodywellfm

Format

data_bodywellfm

A data frame with 106 rows and 2 columns:

Body Satisfaction

numeric

Well-being

numeric

BodyWellM - Ch12 - Body Satisfaction and Well-being data for males from Figure 11.24 left panel

Description

A subset of data_bodywell_fm, reports only for those participants who identified as male. Data is Subjective Wellbeing abd Body Satisfaction.

Usage

data_bodywellm

Format

data_bodywellm

A data frame with 47 rows and 2 columns:

Body Satisfaction

numeric

Well-being

numeric

Campus Involvement - Ch11 - for End-of-Chapter Exercise 11.7

Description

Clinton conducted a survey of college students to determine the extent to which subjective well-being is related to campus involvement (Campus Involvement data set on the book website). Participants completed a measure of subjective well-being (scale from 1 to 5) and a measure of campus involvement (scale from 1 to 5). Participants also reported gender (male or female) and commuter status (resident or commuter). Synthetic data simulated to mimic survey data from a class project.

Usage

data_campus_involvement

Format

data_campus_involvement

A data frame with 113 rows and 6 columns:

ID

integer

Gender

factor

GPA

numeric

CommuterStatus

factor

SWB

numeric

Campus Involvement

numeric

Fictitious data from an unrealistically small HEAT study comparing scores for a single group of students before and after a workshop on climate change.

Description

Fictitious data from an unrealistically small HEAT study comparing scores for a single group of students before and after a workshop on climate change.

Usage

data_chap_8_paired_ex_8.18

Format

data_chap_8_paired_ex_8.18

A data frame with 8 rows and 2 columns:

Before

numeric

After

numeric

Clean moral - Ch07 - from Schnall et al. (2008), Study 1, and Johnson et al. (2014)

Description

Some researchers claim that moral judgments are based not only on rational considerations but also on one's current emotional state. To what extent can recent emotional experiences influence moral judgments? Schnall et al. (2008) examined this question by manipulating feelings of cleanliness and purity and then observing the extent that this changes how harshly participants judge the morality of others. Inscho Study 1, Schnall et al. asked participants to complete a word scramble task with either neutral words (neutral prime) or words related to cleanliness (cleanliness prime). All students then completed a set of moral judgments about controversial scenarios: Moral judgment is the average of six items, each rated on a scale from 0 to 9, with high meaning harsh. The data from this study are in the Clean moral file, which also contains data from a replication by Johnson et al. (2014)

Usage

data_clean_moral

Format

data_clean_moral

A data frame with 208 rows and 4 columns:

Schnall Condition

factor

Schnall Moral judgment

numeric

Johnson Condition

factor

Johnson Moral judgment

numeric

Source

doi:10.1027/1864-9335/a000186

College survey 1 - Ch03 - for End-of-Chapter Exercise 3.3

Description

Data from an additional survey of Dominican University students; reports various psychological and behavioral measures.

Usage

data_college_survey_1

Format

data_college_survey_1

A data frame with 243 rows and 23 columns:

ID

integer

Gender

factor

Gender_Code

factor

Age

integer

Shool_Year

factor

School_Year_Code

factor

Transfer

factor

Transfer_Code

logical

Student_Athlete

factor

Student_Athlete_Code

logical

Wealth_SR

numeric

GPA

numeric

ACT

integer

Subjective_Well_Being

numeric

Positive_Affect

numeric

Negative_Affect

numeric

Relationship_Confidence

numeric

Exercise

numeric

Academic_Motivation_Intrinsic

numeric

Academic_Motivation_Extrinsic

numeric

Academic_Motivation_Amotivation

numeric

Intelligence_Value

numeric

Raven_Score

numeric

College survey 2 - Ch05 - for End-of-Chapter Exercise 5.4

Description

College survey 2 - Ch05 - for End-of-Chapter Exercise 5.4

Usage

data_college_survey_2

Format

data_college_survey_2

A data frame with 138 rows and 17 columns:

ID

integer

Gender

factor

Gender_Code

factor

Age

numeric

Wealth_SR

numeric

School_Year

factor

School_Year_Code

factor

Transfer

factor

Transfer_Code

logical

GPA

numeric

Subjective_Well_Being

numeric

Positive_Affect

numeric

Negative_Affect

numeric

Academic_Engagement

numeric

Religious_Meaning

numeric

Health

numeric

Emotion_Recognition

factor

DamischRCJ - Ch9 - from 6 Damisch studies, and Calin-Jageman and Caldwell (2014)

Description

DamischRCJ - Ch9 - from 6 Damisch studies, and Calin-Jageman and Caldwell (2014)

Usage

data_damischrcj

Format

data_damischrcj

A data frame with 8 rows and 5 columns:

Study

factor

Cohen's d unbiased

numeric

n Control

integer

n Lucky

integer

Research Group

factor

EffronRaj fakenews - Ch8 - from Effron and Raj (2020), v1.1

Description

Synthetic data meant to represent Experiment 1 of Effron & Raj, 2020. 138 U.S. adults, recruited in August 2018 on Prolific Academic, worked online. First, they saw six fake headlines four times, each time being asked to rate how interesting/engaging/ funny/well-written the headline was. This rating task simply ensured that the participants paid some attention to each headline. The stimuli were 12 actual fake-news headlines about American politics, with accompanying photographs. Half appealed to Republicans and half to Democrats. Later, 12 fake headlines were presented one at a time, a random mix of the six Old headlines-those seen before-and six New headlines not seen previously. It was stated very clearly that independent, non-partisan fact-checking had established that all the headlines were not true. Participants first rated, on a 0 (not at all) to 100 (extremely) scale, the degree to which to which they judged it unethical to publish that headline. That's the Unethicality DV. They also rated how likely they would be to share the headline if they saw it posted by an acquaintance on social media; there were three further similar ratings. Finally, they rated how accurate they believed the headline to be.

Usage

data_effronraj_fakenews

Format

data_effronraj_fakenews

A data frame with 138 rows and 5 columns:

ID

factor

UnethOld

numeric

UnethNew

numeric

AccurOld

numeric

AccurNew

numeric

Details

v1.1 – Participant 46 had an invalid negative value for UnethOld; replaced with 0.

Source

doi:10.1177/0956797619887896

Emotion heartrate - Ch8 - from Lakens (2013)

Description

Anger is a powerful emotion. To what extent can feeling angry actually change your heart rate? To investigate, Lakens (2013) asked students to record their heart rate (in beats per minute) at rest before (baseline) and then while recalling a time of intense anger. This is a conceptual replication of a classic study by Ekman et al. (1983). Load the Emotion heartrate data set from the book website.

Usage

data_emotion_heartrate

Format

data_emotion_heartrate

A data frame with 68 rows and 3 columns:

ID

integer

HR_baseline

numeric

HR_anger

numeric

Source

doi:10.1109/T-AFFC.2013.3

Exam Scores - Ch11 - for End-of-Chapter Exercise 11.2

Description

To what extent does initial performance in a class relate to performance on a final exam? First exam and final exam scores for nine students enrolled in an introductory psychology course. Exam scores are percentages, where 0 = no answers correct and 100 = all answers correct. Data is synthetic to represent patterns found in a previous psych stats course.

Usage

data_exam_scores

Format

data_exam_scores

A data frame with 9 rows and 3 columns:

StudentID

factor

First Exam

numeric

Final Exam

numeric

Flag Priming ma - Ch9 - Many Labs replications of Carter et al. (2011)

Description

To what extent does being exposed to the American flag influence political attitudes? One seminal study (Carter et al., 2011) explored this issue by subtly exposing participants either to images of the American flag or to control images. Next, participants were asked about their political attitudes, using a 1-7 rating scale where high scores indicate conservative attitudes. Participants exposed to the flag were found to express substantially more conservative attitudes. The Many Labs project replicated this finding at 25 different locations in the United States.

Usage

data_flag_priming_ma

Format

data_flag_priming_ma

A data frame with 25 rows and 7 columns:

Location

factor

M Flag

numeric

s Flag

numeric

n Flag

integer

M Noflag

numeric

s Noflag

numeric

n Noflag

integer

Source

doi:10.5334/jopd.ad

Gender math IAT - Ch07 - Ithaca and SDSU replications of Nosek et al. (2002)

Description

To what extent do men and women differ in their attitudes towards mathematics? To investigate, Nosek et al. (2002) asked male and female students to complete an Implicit Association Test (IAT)-this is a task designed to measure a participant's implicit (non-conscious) feelings towards a topic. (If you've never heard of the IAT, try it out here: tiny.cc/harvardiat) On this IAT, students were tested for negative feelings towards mathematics and art. Scores reflect the degree to which a student had more negative implicit attitudes about mathematics than art (positive score: more negative feelings about mathematics; 0: same level of negativity to both; negative score: more negative feelings about art). data_gender_math_iat has data from two labs that participated in a large-scale replication of the original study (Klein et al., 2014a, 2014b)

Usage

data_gender_math_iat

Format

data_gender_math_iat

A data frame with 155 rows and 4 columns:

Ithaca gender

factor

Ithaca IAT

numeric

SDSU gender

factor

SDSU IAT

numeric

Source

doi:10.1027/1864-9335/a000178

Gender math IAT ma - Ch9 - Many Labs replications of Nosek et al. (2002)

Description

In EOC Exercise 4 in Chapter 7 we encountered the classic study of Nosek et al. (2002), in which male and female participants completed an Implicit Association Test (IAT) that measured the extent of negative attitudes towards mathematics, compared with art. The study found that women, compared with men, tended to have more negative implicit attitudes towards mathematics. The Many Labs project repeated this study at locations around the world (Klein et al., 2014a, 2014b). Summary data for 30 of these labs are available in Gender math IAT ma. Higher scores indicate more implicit bias against mathematics. See also data_gender_math_iat for raw data from two specific sites from this replication effort.

Usage

data_gender_math_iat_ma

Format

data_gender_math_iat_ma

A data frame with 30 rows and 9 columns:

Location

factor

M Male

numeric

s Male

numeric

n Male

integer

M Female

numeric

s Female

numeric

n Female

integer

USAorNot

factor

Country

factor

Source

doi:10.1027/1864-9335/a000178

Halagappa - Ch14 - from Halagappa et al. (2007)

Description

Could eating much less delay Alzheimer's? If so, that would be great news. Halagappa et al. (2007) investigated the possibility by using a mouse model, meaning they used Alzheimer-prone mice, which were genetically predisposed to develop neural degeneration typical of Alzheimer's. The researchers used six independent groups of mice, three tested in mouse middle age when 10 months old, and three in mouse old age when 17 months. At each age there was a control group of normal mice that ate freely (the NFree10 and NFree17 groups), a group of Alzheimer-prone mice that also ate freely (the AFree10 and AFree17 groups), and another Alzheimer-prone group restricted to 40% less food than normal (the ADiet10 and ADiet17 groups). Table 14.2 lists the factors that define the groups, and group labels. I'll discuss one measure of mouse cognition: the percent time spent near the target of a water maze, with higher values indicating better learning and memory. Table 14.2 reports the means and standard deviations for this measure, and group sizes.

Usage

data_halagappa

Format

data_halagappa

A data frame with 6 rows and 4 columns:

Groups

factor

Mean

numeric

SD

numeric

n

integer

Source

doi:10.1016/j.nbd.2006.12.019

Home Prices - Ch12 - for End-of-Chapter Exercise 12.2

Description

Maybe you're thinking about buying a house after college? Regression can help you hunt for a bargain. Download the Home Prices data set. This file contains real estate listings from 1997 to 2003 in a city in California. Let's explore the extent to which the size of the home (in square meters) predicts the sale price.

Usage

data_home_prices

Format

data_home_prices

A data frame with 300 rows and 8 columns:

MLS

integer

Location

factor

Price

integer

Bedrooms

integer

Bathrooms

integer

Size (m2)

numeric

Status

factor

Status_Code

logical

Kardas Expt 3 - Ch07 - from Kardas and O'Brien (2018), Experiment 3

Description

Suppose you want to change the battery in your phone, cook the perfect souffle, or perform a three-ball juggle. Just as numerous people do every day, you might search online to find a video that shows what to do. Suppose you watch such a video just once. First question: How well would you then predict you could perform the task? Second question: How well would you actually perform the task, the first time you tried? Now suppose you watch the video many times: Again consider the two questions. These questions were investigated in a series of studies by Kardas and O'Brien (2018). Let's first do some quick analyses of Kardas Experiments 3 and 4-let's call them Expt 3 and Expt 4-focusing on the effect of watching a video many times rather than once. In Expt 3, participants first watched a brief video of a person performing the moonwalk. The Low Exposure group watched the video once, the High Exposure group 20 times. Then participants predicted, on a 1 to 10 scale, how well they felt they would be able to perform the moonwalk themselves. Finally, they attempted a single performance of the moonwalk, which was videoed. These videos were rated, on the same 1 to 10 scale, by independent raters.

Usage

data_kardas_expt_3

Format

data_kardas_expt_3

A data frame with 100 rows and 3 columns:

Exposure

factor

Prediction

numeric

Performance

numeric

Source

doi:10.1177/0956797617740646

Kardas Expt 4 - Ch07 - from Kardas and O'Brien (2018), Experiment 4

Description

Suppose you want to change the battery in your phone, cook the perfect souffle, or perform a three-ball juggle. Just as numerous people do every day, you might search online to find a video that shows what to do. Suppose you watch such a video just once. First question: How well would you then predict you could perform the task? Second question: How well would you actually perform the task, the first time you tried? Now suppose you watch the video many times: Again consider the two questions. These questions were investigated in a series of studies by Kardas and O'Brien (2018). Let's first do some quick analyses of Kardas Experiments 3 and 4-let's call them Expt 3 and Expt 4-focusing on the effect of watching a video many times rather than once. Expt 4 was conducted online with participants recruited from Amazon's Mechanical Turk, who are typically more diverse than students. The online task was based on a mirror-drawing game developed by Bob and students (Cusack et al., 2015, tiny.cc/bobmirrortrace). Participants first read a description of the game and the scoring procedure. To play, you use your computer trackpad to trace a target line, as accurately and quickly as you can. The task is tricky because you can see only a mirror image of the path you are tracing with a finger on the trackpad. A running score is displayed. The final score is the percentage match between the target line and the path you traced, so scores can range from 0 to 100

Usage

data_kardas_expt_4

Format

data_kardas_expt_4

A data frame with 270 rows and 4 columns:

Exposure

factor

Prediction

integer

Performance

integer

Time

numeric

Source

doi:10.1177/0956797617740646

Labels flavor - Ch8 - from Floretta-Schiller et al. (2015)

Description

To what extent do brand labels influence perceptions of a product? To investigate, participants were asked to participate in a taste test. All participants were actually given the same grape juice, but one glass was poured from a bottle labeled 'Organic' and the other glass from a bottle labeled 'Generic'. After each tasting (in counterbalanced order), participants were asked to rate how much they enjoyed the juice on a scale from 1 (not at all) to 10 (very much). Participants were also asked to say how much they'd be willing to pay for a large container of that juice on a scale from $1 to $10. Load the Labels flavor data set from the book website. These data were collected as part of a class project by Floretta-Schiller et al. (2015), whose work was inspired by a very clever study looking at the effects of fast-food wrappers on children's enjoyment of food (Robinson et al., 2007).

Usage

data_labels_flavor

Format

data_labels_flavor

A data frame with 51 rows and 6 columns:

ParticipantID

integer

Enjoy_Generic

numeric

Enjoy_Organic

numeric

Pay_Generic

numeric

Pay_Organic

numeric

Suspicious

logical

Latimier 3Groups - Ch14 - 3 groups in Latimier et al. (2019)

Description

The researchers were interested in how different study approaches might impact learning. Working in France, they created three independent groups, each comprising 95 adults. Participants worked online through seven learning modules about DNA. The Reread group worked through a module, then worked through it a second time before going on to the next module. The Quiz group worked through a module, then had to complete a quiz before going on to the next module. The Prequiz group had to work through the quiz before seeing the presentation of a module, then went on to the quiz and presentation of the next module. Participants received feedback and a brief explanation after answering each question in a quiz, and could take as long as they wished to work through each module and quiz. Seven days later, participants completed a final test. data_latimier_3groups is the full data set. To facilitate different student exercises, there are also separate data entities for each group (data_latimier_prequiz, data_latimier_reread, etc.), and for every pair of groups (data_latimier_quiz_prequiz, etc.).

Usage

data_latimier_3groups

Format

data_latimier_3groups

A data frame with 285 rows and 3 columns:

Group

factor

Test%

numeric

Time

numeric

Source

doi:10.7910/DVN/XPYPMF

Latimier Prequiz - Ch03 - Prequiz group in Latimier et al. (2019)

Description

Just the Prequiz group from Latimier et al., 2019 See full details in data_latimier_3_groups

Usage

data_latimier_prequiz

Format

data_latimier_prequiz

A data frame with 95 rows and 3 columns:

Group

factor

Test%

numeric

TIme

numeric

Source

doi:10.7910/DVN/XPYPMF

Latimier Quiz - Ch03 - Quiz group in Latimier et al. (2019)

Description

Just the Quiz group from Latimier et al., 2019 See full details in data_latimier_3_groups

Usage

data_latimier_quiz

Format

data_latimier_quiz

A data frame with 95 rows and 3 columns:

Group

factor

Test%

numeric

Time

numeric

Source

doi:10.7910/DVN/XPYPMF

Latimier Quiz Prequiz - Ch07 - Quiz and Prequiz groups in Latimier et al. (2019)

Description

Just the Quiz (RQ) an Prequiz (QR) groups from Latimier et al., 2019 See full details in data_latimier_3_groups

Usage

data_latimier_quiz_prequiz

Format

data_latimier_quiz_prequiz

A data frame with 190 rows and 3 columns:

Group

factor

Test%

numeric

Time

numeric

Source

doi:10.7910/DVN/XPYPMF

Latimier Reread - Ch03 - Reread group in Latimier et al. (2019)

Description

Just the Reread group from Latimier et al., 2019 See full details in data_latimier_3_groups

Usage

data_latimier_reread

Format

data_latimier_reread

A data frame with 95 rows and 3 columns:

Group

factor

Test%

numeric

Time

numeric

Source

doi:10.7910/DVN/XPYPMF

Latimier Reread Prequiz - Ch07 - Reread and Prequiz groups in Latimier et al. (2019)

Description

Just the Reread (RR) an Prequiz (QR) groups from Latimier et al., 2019 See full details in data_latimier_3_groups

Usage

data_latimier_reread_prequiz

Format

data_latimier_reread_prequiz

A data frame with 190 rows and 3 columns:

Group

factor

Test%

numeric

Time

numeric

Source

doi:10.7910/DVN/XPYPMF

Latimier Reread Quiz - Ch07 - Reread and Quiz groups in Latimier et al. (2019)

Description

Just the Reread Quiz groups from Latimier et al., 2019 See full details in data_latimier_3_groups

Usage

data_latimier_reread_quiz

Format

data_latimier_reread_quiz

A data frame with 190 rows and 3 columns:

Group

factor

Test%

numeric

Time

numeric

Source

doi:10.7910/DVN/XPYPMF

Macnamara r ma - Ch11 - from Macnamara et al. (2014)

Description

Is genius born or made? Could any of us be Michael Jordan or Mozart if we worked sufficiently hard to develop the requisite skills? Meta-analysis of correlations can help answer such questions. The issue here is the extent that practice and effort may be sufficient for achieving the highest levels of expertise. Ericsson et al. (1993) argued that years of effort is what matters most: 'Many characteristics once believed to reflect innate talent are actually the result of intense practice extended for a minimum of 10 years' (p. 363). This view was enormously popularized by Malcolm Gladwell (2008), who argued in his book Outliers that 10,000 hours of focused practice is the key to achieving expertise. However, this view is now being challenged, with one important contribution being a large meta-analysis of correlations between amount of intense practice and level of achievement: Macnamara et al. (2014) combined 157 correlations reported in a wide range of fields, from sports to music and education, and found correlation of only r = .35 (.30, .39). Table 11.1 shows the 16 main correlations for music, from Macnamara et al. (2014).

Usage

data_macnamara_r_ma

Format

data_macnamara_r_ma

A data frame with 16 rows and 4 columns:

Study

factor

r

numeric

N

integer

Instrument_Type

factor

Source

doi:10.1177/0956797614535810

McCabeMichael brain - Ch9 - from Michael et al. (2013)

Description

You've probably seen cross sections of the brain with brightly colored areas indicating which brain regions are most active during a particular type of cognition or emotion. Search online for fMRI (functional magnetic resonance imaging) brain scans to see such pictures and learn how they are made. They can be fascinating-are we at last able to see how thinking works? In 2008, McCabe and Castel published studies that investigated how adding a brain picture might alter judgments of the credibility of a scientific article. For one group of participants, an article was accompanied by a brain image that was irrelevant to the article. For a second, independent group, there was no image. Participants read the article, then gave a rating of the statement 'The scientific reasoning in the article made sense'. The response options were 1 (strongly disagree), 2 (disagree), 3 (agree), and 4 (strongly agree). The researchers reported that mean ratings were higher with a brain picture than without, but that the difference was small. It seemed that an irrelevant brain picture may have some, but only a small influence. The authors drew appropriately cautious conclusions, but the result quickly attracted attention and there were many media reports that greatly overstated it. At least according to the popular media, it seemed that adding a brain picture made any story convincing. Search on 'McCabe seeing is believing', or similar, to find media reports and blog posts. Some warned readers to watch out for brain pictures, which, they said, can trick you into believing things that aren't true. The result intrigued some New Zealander colleagues of mine who discovered that, despite its wide recognition, the finding hadn't been replicated. They ran replication studies using the materials used by the original researchers, and found generally small ESs. I joined the team at the data analysis stage and the research was published (Michael et al., 2013). I'll discuss here a meta-analysis of two of the original studies and eight replications by our team. The studies were sufficiently similar for meta-analysis, especially considering that all the Michael studies were designed to have many features that matched the original studies. This data set does not include two additional critique studies run by the Michael team. See also data_mccabemichael_brain2

Usage

data_mccabemichael_brain

Format

data_mccabemichael_brain

A data frame with 10 rows and 9 columns:

Study name

factor

M No Brain

numeric

s No Brain

numeric

n No Brain

numeric

M Brain

numeric

s Brain

numeric

n Brain

numeric

SimpleCritique

factor

Research group

factor

Source

doi:10.3758/s13423-013-0391-6

McCabeMichael brain2 - Ch9 - from Michael et al. (2013)

Description

Same as data_mccabemichael_brain but includes two additional critique studies run by the Michael team.

Usage

data_mccabemichael_brain2

Format

data_mccabemichael_brain2

A data frame with 12 rows and 9 columns:

Study name

factor

M No Brain

numeric

s No Brain

numeric

n No Brain

numeric

M Brain

numeric

s Brain

numeric

n Brain

numeric

SimpleCritique

factor

Research group

factor

Source

doi:10.3758/s13423-013-0391-6

MeditationBrain - Ch15 - from Holzel et al. (2011)

Description

My example is a well-known study of mindfulness meditation by Holzel et al. (2011). People who wanted to reduce stress, and were not experienced meditators, were assigned to a Meditation (n = 16) or a Control (n = 17) group. The Meditation group participated in 8 weeks of intensive training and practice of mindfulness meditation. The researchers used a questionnaire to assess a range of emotional and cognitive variables both before (Pretest) and after (Posttest) the 8-week period. All assessment was conducted while the participants were not meditating. The study is notable for including brain imaging to assess possible changes in participants' brains from Pretest to Posttest. The researchers measured gray matter concentration, which increases in brain regions that experience higher and more frequent activation. The researchers expected that the hippocampus may be especially responsive to meditation because it has been implicated in the regulation of emotion, arousal, and general responsiveness. They therefore included in their planned analysis the assessment of any changes to gray matter concentration in the hippocampus.

Usage

data_meditationbrain

Format

data_meditationbrain

A data frame with 33 rows and 7 columns:

Pretest

numeric

Posttest

numeric

Group

factor

ControlPre

numeric

ControlPost

numeric

MeditationPre

numeric

MeditationPost

numeric

Source

doi:10.1007/978-94-007-2079-4_9

OrganicMoral - Ch14 - from Eskine (2013)

Description

To what extent might choosing organic foods make us morally smug? To investigate, Eskine (2013) asked participants to rate images of organic food, neutral (control) food, or comfort food. Next, under the guise of a different study, all participants completed a moral judgment scale in which they read different controversial scenarios and rated how morally wrong they judged them to be (scale of 1-7, high judgments mean more wrong). Table 14.7 shows summary data, which are also available in the first four variables in the OrganicMoral file. In that file you can see two further variables, which report full data-we'll come to these shortly. Here we use the summary data. After the results of Eskine (2013) were published, Moery and Calin-Jageman (2016) conducted a series of close replications. We obtained original materials from Eskine, piloted the procedure, and preregistered our sampling and analysis plan. The OSF page, osf.io/atkn7, has all the details. The data from one of these close replications are in the last two variables of the OrganicMoral file. For this replication study, group names are in the variable ReplicationGroup and moral judgments in MoralJudgment. (You may need to scroll right to see these variables.)

Usage

data_organicmoral

Format

data_organicmoral

A data frame with 106 rows and 6 columns:

Group

factor

Mean

numeric

SD

numeric

N

integer

ReplicationGroup

factor

MoralJudgmentment

numeric

Source

doi:10.1177/1948550616639649

% transcription scores from pen and laptop group of Meuller et al., 2014

Description

% transcription scores from pen and laptop group of Meuller et al., 2014

Usage

data_penlaptop1

Format

data_penlaptop1

A data frame with 65 rows and 2 columns:

condition

factor

transcription

numeric

Source

doi:10.1177/0956797614524581

PowerPerformance ma - Ch9 - from Burgmer and Englich (2012), and Cusack et al. (2015)

Description

To what extent could feeling powerful affect your performance at motor skills? To investigate, Burgmer and Englich (2012) assigned German participants to either power or control conditions and then asked them to play golf (Experiment 1) or darts (Experiment 2). They found that participants manipulated to feel powerful performed substantially better than those in the control condition. To study this finding further, Cusack et al. (2015) conducted five replications in the United States. Across these replications they tried different ways of manipulating power, different types of tasks (golf, mirror tracing, and a cognitive task), different levels of difficulty, and different types of participant pools (undergraduates and online). Summary data from all seven studies are available in PowerPerformance ma.

Usage

data_powerperformance_ma

Format

data_powerperformance_ma

A data frame with 8 rows and 12 columns:

StudyName

factor

Country

factor

Population

factor

Difficulty

factor

Task

factor

M Control

numeric

s Control

numeric

M Power

numeric

s Power

numeric

Cohen d unb

numeric

n Control

integer

n Power

integer

Source

doi:10.1371/journal.pone.0140806

RattanMotivation - Ch14 - from Rattan et al. (2012)

Description

How do you think you would react to feedback that gave encouragement and reassurance, or, instead, encouragement and challenge? Carol Dweck and her colleagues have investigated many such questions about how people respond to different types of feedback. My next example comes from Dweck's research group and illustrates data analysis that starts with the full data, rather than only summary statistics. Rattan et al. (2012) asked their college student participants to imagine they were undertaking a mathematics course and had just received a low score (65%) on the first test of the year. Participants were assigned randomly into three groups, which received different feedback along with the low score. The Comfort group received positive encouragement and also reassurance, the Challenge group received positive encouragement and also challenge, and the Control group received just the positive encouragement. Participants then responded to a range of questions about how they felt about the course and their professor. I'll discuss data for their ratings of their own motivation toward mathematics, made after they had received the feedback.

Usage

data_rattanmotivation

Format

data_rattanmotivation

A data frame with 54 rows and 2 columns:

Group

factor

Motivation

numeric

Source

doi:10.1016/j.jesp.2011.12.012

ReligionSharing - Ch14 - RETRACTED DATA used in End-of-Chapter Exercise 14.3

Description

To what extent is a religious upbringing related to prosocial behavior in childhood? To investigate, a large international sample of children was asked to play a game in which they were given 10 stickers but then asked if they would give some of these stickers away to another child who had not been able to be tested that day. The number of stickers donated was considered a measure of altruistic sharing. In addition, the parents of each child reported the family's religion. Summary data provided. THIS STUDY HAS BEEN RETRACTED DUE TO AN ERRONEOUS ANALYSIS - THE DATASET WILL BE REMOVED FROM FUTURE VERSIONS OF ESCI AND THE BOOK

Usage

data_religionsharing

Format

data_religionsharing

A data frame with 3 rows and 4 columns:

Group

factor

Mean

numeric

SD

numeric

N

integer

Source

doi:10.1016/j.cub.2015.09.056

Religious belief - Ch03 - for End-of-Chapter Exercise 3.5

Description

Let's look at some data about religious beliefs. The Religious belief file has data from a large online survey in which participants were asked to report, on a scale from 0 to 100, their belief in the existence of God. Age was also reported.

Usage

data_religious_belief

Format

data_religious_belief

A data frame with 213 rows and 3 columns:

Response_ID

character

Belief_in_God

factor

Age

integer

SelfExplain - Ch15 - from McEldoon et al. (2013)

Description

Self-explaining is a learning strategy where students write or say their own explanations of the material they are studying. Self-explaining has generally been found to be more effective than standard studying, but it may also take more time. This raises the question of whether it's the study strategy or the extra time that benefits learning. To explore this issue, grade school children took a pretest of mathematics conceptual knowledge, studied mathematics problems, and then took a similar posttest (McEldoon et al., 2013). Participants were randomly assigned to one of two study conditions: normal study + more practice (More Practice group), or self-explaining (Self-Explain group). The first condition was intended to make time spent learning to be similar for the two groups. You can find part of the data from this study in SelfExplain, the scores being percent correct.

Usage

data_selfexplain

Format

data_selfexplain

A data frame with 52 rows and 4 columns:

Student ID

factor

Condition

factor

Pretest

numeric

Posttest

numeric

Source

doi:10.1111/j.2044-8279.2012.02083.x

SimmonsCredibility - Ch14 - from Simmons and Nelson (2020)

Description

You're excited! Your company has developed a wonderful new weight-loss program, and now it's your job to develop the ad campaign. Should you choose a BeforeAfter pair of pictures, as in Figure 14.1, top panel? Or might a Progressive sequence of pictures of the same person, as in the bottom panel, be more effective? Pause, think, and discuss. Which would you choose, and why? You might think that BeforeAfter is simpler and more dramatic. On the other hand, Progressive highlights the steady improvement that you claim the program will deliver. You're probably not surprised to learn that BeforeAfter is used often and has long been a favorite of the advertising industry, whereas Progressive is used only rarely. Luca Cian and colleagues (Cian et al., 2020) were curious to know the extent to which BeforeAfter is actually more effective, appealing, and credible than Progressive, or, indeed, whether Progressive might score more highly. They reported seven studies of various aspects of that question. I'll focus on their Study 2, in which they used three independent groups to compare all three conditions illustrated in Figure 14.1. The BeforeAfterInfo condition, in the middle panel, comprises three BeforeAfter pairs, thus providing extra information about the before and after endpoints. The researchers included this condition in case any advantage of Progressive might stem simply from having more images, rather than because it illustrates a clear progressive sequence. They randomly assigned 213 participants from MTurk to one of the three groups. Participants were asked to 'imagine that you have decided to lose some weight', then saw one of the three ads for a weight loss program called MRMDiets. They then answered the question 'How would you evaluate MRMDiets?' by choosing a 1-7 response on several scales, including Unlikeable-Likable, Ineffective-Effective, and Not credible-Credible. The researchers averaged six such scores to give an overall Credibility score, on the 1-7 scale, with 7 being the most credible. Simmons and Nelson (2020) were sufficiently intrigued to carry out two substantial very close replications. With the cooperation of the original researchers, they used the same materials and procedure. They used much larger groups and preregistered their research plan, including data analysis plan. I'll focus on their first replication, in which 761 participants from MTurk were randomized to the three groups.

Usage

data_simmonscredibility

Format

data_simmonscredibility

A data frame with 3 rows and 4 columns:

Groups

factor

Mean

numeric

SD

numeric

n

integer

Source

http://datacolada.org/94

Sleep Beauty - Ch11 - for End-of-Chapter Exercise 11.6

Description

Is there really such a thing as beauty sleep? To investigate, researchers decided to examine the extent to which sleep relates to attractiveness. Each of 70 college students self-reported the amount of sleep they had the night before. In addition, a photograph was taken of each participant and rated for attractiveness on a scale from 1 to 10 by two judges of the opposite gender. The average rating score was used. You can download this data set (Sleep Beauty) from the book website.

Usage

data_sleep_beauty

Format

data_sleep_beauty

A data frame with 70 rows and 2 columns:

Sleep (hours)

numeric

Rated_Attractiveness

numeric

SmithRecall - Ch15 - from Smith et al. (2016)

Description

SmithRecall - Ch15 - from Smith et al. (2016)

Usage

data_smithrecall

Format

data_smithrecall

A data frame with 120 rows and 6 columns:

ID

integer

Study Technique

factor

Stress Status

factor

%Recalled

numeric

Items Recalled

numeric

Group

factor

Stickgold - Ch06 - from Stickgold et al. (2000)

Description

Stickgold et al. (2000) found that, remarkably, performance on a visual discrimination task actually improved over the 48-96 hours after initial training, even without practice during that time. However, what if participants were sleep deprived during that period? They trained 11 participants in that new skill, then all were sleep deprived. The data were (-14.7, -10.7, -10.7, 2.2, 2.4, 4.5, 7.2, 9.6, 10, 21.3, 21.8)-or download the Stickgold data set from the book website. The data are the changes in performance scores from immediately after training to after the night without sleep: 0 represents no change, positive scores represent improvement, and negative scores represent decline. Data set courtesy of DataCrunch (tiny.cc/Stickgold)

Usage

data_stickgold

Format

data_stickgold

A data frame with 11 rows and 3 columns:

Sleep deprived

numeric

B

factor

C

factor

Source

https://www.statcrunch.com/app/index.html?dataid=1053539

StudyStrategies - Ch14 - from O'Reilly et al. (1998)

Description

To what extent does study strategy influence learning? To investigate, psychology students were randomly assigned to three groups and asked to learn biology facts using one of three different strategies: a) Self-Explain (explaining for each fact what new knowledge is gained and how it relates to what is already known), b) Elab Interrogation (elaborative interrogation: stating for each fact why it makes sense), or c) Repetition Control (stating each fact over and over). After studying, students took a 25-point fill-the-blank test (O'Reilly et al., 1998)

Usage

data_studystrategies

Format

data_studystrategies

A data frame with 3 rows and 10 columns:

Group

factor

TestMean

numeric

TestSD

numeric

TestN

integer

PrevKnowMean

numeric

PrevKnowSD

numeric

PrevKnowN

integer

EaseUseMean

numeric

EaseUseSD

numeric

EaseUseN

integer

Source

doi:10.1006/ceps.1997.0977

Thomason 1 - Ch11 - from Thomason 1

Description

Summary data from an unpublished study by Neil Thomason and colleagues, who were interested in ways to enhance students' critical thinking. They were investigating argument mapping, which is a promising way to use diagrams to represent the structure of arguments. Students in their study completed an established test of critical thinking (the Pretest), then a critical thinking course based on argument mapping, then a second version of the test (the Posttest).

Usage

data_thomason_1

Format

data_thomason_1

A data frame with 12 rows and 3 columns:

Participant ID

factor

Pretest

numeric

Posttest

numeric

VideogameAggression - Ch15 - from Hilgard (2015)

Description

Video games can be violent and they can also be challenging. To what extent might these factors cause aggressive behavior? To explore, Hilgard (2015) asked male participants to play one of four versions of a video game for 15 minutes. The game was customized so that it could vary in violence (shooting zombies or helping aliens) and difficulty (targets controlled by tough AI or dumb AI). After the game, players were provoked by being given an insulting evaluation by a confederate. Participants then got to decide how long the confederate should hold their hand in painfully cold ice water (between 0 and 80 seconds), and this was taken as a measure of aggressive behavior. You can find the materials and analysis plan for this study on the Open Science Framework: osf. io/cwenz. This is a simplified version of the full data set.

Usage

data_videogameaggression

Format

data_videogameaggression

A data frame with 223 rows and 3 columns:

Violence

factor

Difficulty

factor

Agression

numeric

Source

doi:10.1177/0956797619829688

Add a difference axis to the x axis of an esci forest plot

Description

esci_plot_difference_axis_x can be used to redraw the difference axis from a forest plot created with plot_meta. You must pass the plot returned from plot_meta and the effect size table containing the estimated difference.

Usage

esci_plot_difference_axis_x(
  myplot,
  difference_table,
  dlim = c(NA, NA),
  d_n.breaks = NULL,
  d_lab = NULL
)

Arguments

Estimates for a continuous variable with no grouping (single-group design)

Description

estimate_magnitude is suitable for a single group design with a continuous outcome variable. It estimates the population mean and population median (raw data only) with confidence intervals. You can pass raw data or summary data.

Usage

estimate_magnitude(
  data = NULL,
  outcome_variable = NULL,
  mean = NULL,
  sd = NULL,
  n = NULL,
  outcome_variable_name = "My outcome variable",
  conf_level = 0.95,
  save_raw_data = TRUE
)

Arguments

Details

Reach for this function in place of a one-sample t-test or z-test.

Once you generate an estimate with this function, you can visualize it with plot_magnitude().

If you want to compare your sample to a known value or reference, then use estimate_mdiff_one().

The estimated mean is from statpsych::ci.mean1() (renamed ci.mean as of statpsych 1.6).

The estimated median is from statpsych::ci.median1() (renamed ci.median as of statpsych 1.6)

Value

Returns an object of class esci_estimate

• overview

  • outcome_variable_name -

  • mean -

  • mean_LL -

  • mean_UL -

  • median -

  • median_LL -

  • median_UL -

  • sd -

  • min -

  • max -

  • q1 -

  • q3 -

  • n -

  • missing -

  • df -

  • mean_SE -

  • median_SE -

• es_mean

  • outcome_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • df -

  • ta_LL -

  • ta_UL -

• es_median

  • outcome_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • df -

  • ta_LL -

  • ta_UL -

• raw_data

  • grouping_variable -

  • outcome_variable -

Examples

# From raw data
data("data_penlaptop1")

estimate_from_raw <- esci::estimate_magnitude(
  data = data_penlaptop1[data_penlaptop1$condition == "Pen", ],
    outcome_variable = transcription
)

# To visualize the estimate
myplot_from_raw <- esci::plot_magnitude(
  estimate_from_raw,
  effect_size = "median"
)


# From summary data
mymean <- 24.5
mysd <- 3.65
myn <- 40

estimate_from_summary <- esci::estimate_magnitude(
  mean = mymean,
  sd = mysd,
  n = myn
)

# To visualize the estimate
myplot_from_summary <- esci::plot_magnitude(
  estimate_from_summary,
  effect_size = "mean"
)

Estimates for a 2x2 between-subjects design with a continuous outcome variable

Description

Returns object estimate_mdiff_2x2_between is suitable for a 2x2 between-subjects design with a continuous outcome variable. It estimates each main effect, the simple effects for the first factor, and the interaction. It can express these estimates as mean differences, standardized mean differences (Cohen's d), and as median differences (raw data only). You can pass raw data or or summary data (summary data does not return medians).

Usage

estimate_mdiff_2x2_between(
  data = NULL,
  outcome_variable = NULL,
  grouping_variable_A = NULL,
  grouping_variable_B = NULL,
  means = NULL,
  sds = NULL,
  ns = NULL,
  grouping_variable_A_levels = NULL,
  grouping_variable_B_levels = NULL,
  outcome_variable_name = "My outcome variable",
  grouping_variable_A_name = "A",
  grouping_variable_B_name = "A",
  conf_level = 0.95,
  assume_equal_variance = FALSE,
  save_raw_data = TRUE
)

Arguments

Details

Reach for this function in place of a 2x2 between-subjects ANOVA.

Once you generate an estimate with this function, you can visualize it with plot_mdiff() and you can visualize the interaction specifically with plot_interaction(). You can test hypotheses with test_mdiff().

The estimated mean differences are from statpsych::ci.2x2.mean.bs().

The estimated SMDs are from statpsych::ci.2x2.stdmean.bs().

The estimated median differences are from statpsych::ci.2x2.median.bs()

Value

Returns object of class esci_estimate

• es_mean_difference

  • type -

  • outcome_variable_name -

  • grouping_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • df -

  • ta_LL -

  • ta_UL -

  • effect_type -

  • effects_complex -

• es_median_difference

  • type -

  • outcome_variable_name -

  • grouping_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • ta_LL -

  • ta_UL -

  • effect_type -

  • effects_complex -

• es_smd

  • outcome_variable_name -

  • grouping_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • numerator -

  • denominator -

  • SE -

  • df -

  • d_biased -

  • effect_type -

  • effects_complex -

• overview

  • outcome_variable_name -

  • grouping_variable_name -

  • grouping_variable_level -

  • mean -

  • mean_LL -

  • mean_UL -

  • median -

  • median_LL -

  • median_UL -

  • sd -

  • min -

  • max -

  • q1 -

  • q3 -

  • n -

  • missing -

  • df -

  • mean_SE -

  • median_SE -

• raw_data

  • grouping_variable -

  • outcome_variable -

  • grouping_variable_A -

  • grouping_variable_B -

Examples

data("data_videogameaggression")

estimates_from_raw <- esci::estimate_mdiff_2x2_between(
  esci::data_videogameaggression,
  Agression,
  Violence,
  Difficulty
)

# To visualize the estimated mean difference for the interaction
myplot_from_raw <- esci::plot_mdiff(
  estimates_from_raw$interaction,
  effect_size = "median"
)

# To conduct a hypothesis test on the mean difference
res_htest_from_raw <- esci::test_mdiff(
  estimates_from_raw$interaction,
  effect_size = "median"
)


# From summary data
means <- c(1.5, 1.14, 1.38, 2.22)
sds <- c(1.38, .96,1.5, 1.68)
ns <- c(26, 26, 25, 26)
grouping_variable_A_levels <- c("Evening", "Morning")
grouping_variable_B_levels <- c("Sleep", "No Sleep")

estimates_from_summary <- esci::estimate_mdiff_2x2_between(
  means = means,
  sds = sds,
  ns = ns,
  grouping_variable_A_levels = grouping_variable_A_levels,
  grouping_variable_B_levels = grouping_variable_B_levels,
  grouping_variable_A_name = "Testing Time",
  grouping_variable_B_name = "Rest",
  outcome_variable_name = "False Memory Score",
  assume_equal_variance = TRUE
)

# To visualize the estimated mean difference for the interaction
plot_mdiff_interaction <- esci::plot_mdiff(
  estimates_from_summary$interaction,
  effect_size = "mean"
)

# To visualize the interaction as a line plot
plot_interaction_line <- esci::plot_interaction(estimates_from_summary)

# Same but with fan effect representing each simple-effect CI
plot_interaction_line_CI <- esci::plot_interaction(
  estimates_from_summary,
  show_CI = TRUE
)

# To conduct a hypothesis test on the mean difference
res_htest_from_raw <- esci::test_mdiff(
  estimates_from_summary$interaction,
  effect_size = "mean"
)

Estimates for a 2x2 mixed factorial design with a continuous outcome variable

Description

Returns object estimate_mdiff_2x2_mixed is suitable for a 2x2 mixed-factorial design with a continuous outcome variable. It estimates each main effect, the simple effects for the repeated-measures factor, and the interaction. It can express these estimates as mean differences, median difference, or standardized mean differences. This function accepts raw data only.

Usage

estimate_mdiff_2x2_mixed(
  data,
  outcome_variable_level1,
  outcome_variable_level2,
  grouping_variable,
  outcome_variable_name = "My outcome variable",
  repeated_measures_name = "Time",
  conf_level = 0.95,
  save_raw_data = TRUE
)

Arguments

Details

Reach for this function in place of a 2x2 mixed-factorial ANOVA.

Once you generate an estimate with this function, you can visualize it with plot_mdiff() and you can visualize the interaction specifically with plot_interaction(). You can test hypotheses with test_mdiff().

The estimated mean differences are from statpsych::ci.2x2.mean.mixed().

Value

Returns object of class esci_estimate

• es_mean_difference

  • type -

  • outcome_variable_name -

  • grouping_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • df -

  • ta_LL -

  • ta_UL -

  • effect_type -

  • effects_complex -

  • t -

  • p -

• es_smd

  • outcome_variable_name -

  • grouping_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • numerator -

  • denominator -

  • SE -

  • df -

  • d_biased -

  • effect_type -

  • effects_complex -

• overview

  • outcome_variable_name -

  • grouping_variable_name -

  • grouping_variable_level -

  • mean -

  • mean_LL -

  • mean_UL -

  • median -

  • median_LL -

  • median_UL -

  • sd -

  • min -

  • max -

  • q1 -

  • q3 -

  • n -

  • missing -

  • df -

  • mean_SE -

  • median_SE -

• raw_data

  • grouping_variable -

  • outcome_variable -

  • grouping_variable_A -

  • grouping_variable_B -

  • paired -

Examples

# From raw data (summary data mode not available for this function)
example_data <- data.frame(
  pretest = c(
    19, 18, 19, 20, 17, 16, 16, 10, 12,  9, 13, 15
  ),
  posttest = c(
    18, 19, 20, 17, 20, 16, 19, 16, 16, 14, 16, 18
  ),
  condition = as.factor(
    c(
      rep("Control", times = 6),
      rep("Treated", times = 6)
    )
  )
)

estimates <- esci::estimate_mdiff_2x2_mixed(
  data = example_data,
  outcome_variable_level1 = pretest,
  outcome_variable_level2 = posttest,
  grouping_variable = condition,
  repeated_measures_name = "Time"
)
# To visualize the estimated mean difference for the interaction
myplot <- esci::plot_mdiff(estimates$interaction, effect_size = "mean")

# Line-plot of the interaction with fan effect representing each simple-effect CI
plot_interaction_line_CI <- esci::plot_interaction(
  estimates,
  show_CI = TRUE
)

# To conduct a hypothesis test
res_htest_from_raw <- esci::test_mdiff(
  estimates$interaction,
  effect_size = "mean"
 )

Estimates for a multi-group design with a continuous outcome variable

Description

Returns object estimate_mdiff_ind_contrast is suitable for a multi-group design (between subjects) with a continuous outcome variable. It accepts a user-defined set of contrast weights that allows estimation of any 1-df contrast. It can express estimates as mean differences, standardized mean differences (Cohen's d) or median differences (raw data only). You can pass raw data or summary data.

Usage

estimate_mdiff_ind_contrast(
  data = NULL,
  outcome_variable = NULL,
  grouping_variable = NULL,
  means = NULL,
  sds = NULL,
  ns = NULL,
  contrast = NULL,
  grouping_variable_levels = NULL,
  outcome_variable_name = "My outcome variable",
  grouping_variable_name = "My grouping variable",
  conf_level = 0.95,
  assume_equal_variance = FALSE,
  save_raw_data = TRUE
)

Arguments

Details

Reach for this function in place of a one-way ANOVA.

Once you generate an estimate with this function, you can visualize it with plot_mdiff() and you can test hypotheses with test_mdiff().

The estimated mean differences are from statpsych::ci.lc.mean.bs().

The estimated SMDs are from CI_smd_ind_contrast() which relies on statpsych::ci.lc.stdmean.bs() unless there are only 2 groups.

The estimated median differences are from statpsych::ci.lc.median.bs()

Value

Returns object of class esci_estimate

• es_mean_difference

  • type -

  • outcome_variable_name -

  • grouping_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • df -

  • ta_LL -

  • ta_UL -

• es_median_difference

  • type -

  • outcome_variable_name -

  • grouping_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • ta_LL -

  • ta_UL -

• es_smd

  • outcome_variable_name -

  • grouping_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • numerator -

  • denominator -

  • SE -

  • df -

  • d_biased -

• overview

  • outcome_variable_name -

  • grouping_variable_name -

  • grouping_variable_level -

  • mean -

  • mean_LL -

  • mean_UL -

  • median -

  • median_LL -

  • median_UL -

  • sd -

  • min -

  • max -

  • q1 -

  • q3 -

  • n -

  • missing -

  • df -

  • mean_SE -

  • median_SE -

• raw_data

  • grouping_variable -

  • outcome_variable -

Examples

# From raw data
data("data_rattanmotivation")

estimate_from_raw <- esci::estimate_mdiff_ind_contrast(
  esci::data_rattanmotivation,
  Motivation,
  Group,
  contrast = c("Challenge" = 1, "Control" = -1/2, "Comfort" = -1/2)
)

# To visualize the estimate
myplot_from_raw <- esci::plot_mdiff(
  estimate_from_raw,
  effect_size = "median"
)

# To conduct a hypothesis test
res_htest_from_raw <- esci::test_mdiff(
  estimate_from_raw,
  effect_size = "median"
)


# From summary data
data("data_halagappa")

estimate_from_summary <- estimate_mdiff_ind_contrast(
  means = data_halagappa$Mean,
  sds = data_halagappa$SD,
  ns = data_halagappa$n,
  grouping_variable_levels = as.character(data_halagappa$Groups),
  assume_equal_variance = TRUE,
  contrast = c(
    "NFree10" = 1/3,
    "AFree10" = 1/3,
    "ADiet10" = -1/3,
    "NFree17" = -1/3,
    "AFree17" = 1/3,
    "ADiet17" = -1/3
  ),
  grouping_variable_name = "Diet",
  outcome_variable_name = "% time near target"
)

# To visualize the estimate
myplot <- esci::plot_mdiff(estimate_from_summary, effect_size = "mean")

# To conduct a hypothesis test
res_htest_from_raw <- esci::test_mdiff(
  estimate_from_summary,
  effect_size = "mean"
)

Estimates for a single-group design with a continuous outcome variable compared to a reference or population value

Description

Returns object estimate_mdiff_one is suitable for a single-group design with a continuous outcome variable that is compared to a reference or population value. It can express estimates as mean differences, standardized mean differences (Cohen's d) or median differences (raw data only). You can pass raw data or summary data.

Usage

estimate_mdiff_one(
  data = NULL,
  outcome_variable = NULL,
  comparison_mean = NULL,
  comparison_sd = NULL,
  comparison_n = NULL,
  reference_mean = 0,
  outcome_variable_name = "My outcome variable",
  conf_level = 0.95,
  save_raw_data = TRUE
)

Arguments

Details

Reach for this function in place of a z-test or one-sample t-test.

Once you generate an estimate with this function, you can visualize it with plot_mdiff() and you can test hypotheses with test_mdiff().

The estimated mean differences are from statpsych::ci.mean1() (renamed ci.mean as of statpsych 1.6).

The estimated SMDs are from CI_smd_one().

The estimated median differences are from statpsych::ci.median1() (renamed ci.median as of statpsych 1.6)

Value

Returns object of class esci_estimate

• overview

  • outcome_variable_name -

  • mean -

  • mean_LL -

  • mean_UL -

  • median -

  • median_LL -

  • median_UL -

  • sd -

  • min -

  • max -

  • q1 -

  • q3 -

  • n -

  • missing -

  • df -

  • mean_SE -

  • median_SE -

• es_mean

  • outcome_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • df -

  • ta_LL -

  • ta_UL -

• es_median

  • outcome_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • df -

  • ta_LL -

  • ta_UL -

• raw_data

  • grouping_variable -

  • outcome_variable -

• es_mean_difference

  • outcome_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • df -

  • ta_LL -

  • ta_UL -

  • type -

• es_median_difference

  • outcome_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • df -

  • ta_LL -

  • ta_UL -

  • type -

• es_smd

  • outcome_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • numerator -

  • denominator -

  • SE -

  • df -

  • d_biased -

Examples

# From raw data
data("data_penlaptop1")
estimate_from_raw <- esci::estimate_mdiff_one(
  data = data_penlaptop1[data_penlaptop1$condition == "Pen", ],
  outcome_variable = transcription,
  reference_mean = 10
)

# To visualize the mean difference estimate
myplot_from_raw <- esci::plot_mdiff(estimate_from_raw, effect_size = "mean")

# To conduct a hypothesis test
res_htest_from_raw <- esci::test_mdiff(
  estimate_from_raw,
  effect_size = "mean",
  rope = c(-2, 2)
)


# From summary data
mymean <- 12.09
mysd <- 5.52
myn <- 103

estimate_from_summary <- esci::estimate_mdiff_one(
  comparison_mean = mymean,
  comparison_sd = mysd,
  comparison_n = myn,
  reference_mean = 12
)

# To visualize the estimate
myplot_from_sumary <- esci::plot_mdiff(
  estimate_from_summary,
  effect_size = "mean"
)

# To conduct a hypothesis test
res_htest_from_summary <- esci::test_mdiff(
  estimate_from_summary,
  effect_size = "mean",
  rope = c(-2, 2)
)

Estimates for a repeated-measures study with two measures of a continuous variable

Description

Returns object estimate_mdiff_paired is suitable for a simple paired design with a continuous outcome variable. It provides estimates and CIs for the population mean difference between the repeated measures, the standardized mean difference (SMD; Cohen's d) between the repeated measures, and the median difference between the repeated measures (raw data only). You can pass raw data or summary data.

Usage

estimate_mdiff_paired(
  data = NULL,
  comparison_measure = NULL,
  reference_measure = NULL,
  comparison_mean = NULL,
  comparison_sd = NULL,
  reference_mean = NULL,
  reference_sd = NULL,
  n = NULL,
  correlation = NULL,
  comparison_measure_name = "Comparison measure",
  reference_measure_name = "Reference measure",
  conf_level = 0.95,
  save_raw_data = TRUE
)

Arguments

Details

Reach for this function in place of a paired-samples t-test.

Once you generate an estimate with this function, you can visualize it with plot_mdiff() and you can test hypotheses with test_mdiff().

The estimated mean differences are from statpsych::ci.mean.ps().

The estimated SMDs are from CI_smd_ind_contrast().

The estimated median differences are from statpsych::ci.median.ps().

Value

Returns object of class esci_estimate

• overview

  • outcome_variable_name -

  • mean -

  • mean_LL -

  • mean_UL -

  • median -

  • median_LL -

  • median_UL -

  • sd -

  • min -

  • max -

  • q1 -

  • q3 -

  • n -

  • missing -

  • df -

  • mean_SE -

  • median_SE -

• es_mean_difference

  • type -

  • comparison_measure_name -

  • reference_measure_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • df -

  • ta_LL -

  • ta_UL -

• es_smd

  • comparison_measure_name -

  • reference_measure_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • numerator -

  • denominator -

  • SE -

  • d_biased -

  • df -

• es_r

  • x_variable_name -

  • y_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • n -

  • df -

  • ta_LL -

  • ta_UL -

• es_median_difference

  • type -

  • comparison_measure_name -

  • reference_measure_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • ta_LL -

  • ta_UL -

• es_mean_ratio

  • comparison_measure_name -

  • reference_measure_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • comparison_mean -

  • reference_mean -

• es_median_ratio

  • comparison_measure_name -

  • reference_measure_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • comparison_median -

  • reference_median -

• raw_data

  • comparison_measure -

  • reference_measure -

Examples

# From raw data
data("data_thomason_1")

estimate_from_raw <- esci::estimate_mdiff_paired(
  data = esci::data_thomason_1,
  comparison_measure = Posttest,
  reference_measure = Pretest
)

# To visualize the estimated median difference (raw data only)
myplot_from_raw <- esci::plot_mdiff(estimate_from_raw, effect_size = "median")

# To conduct a hypothesis test
res_htest_from_raw <- esci::test_mdiff(
  estimate_from_raw,
  effect_size = "median",
  rope = c(-2, 2)
)


sd1 <- 4.28
sd2 <- 3.4
sdiff <- 2.13

cor <- (sd1^2 + sd2^2 - sdiff^2) / (2*sd1*sd2)

estimate_from_summary <- esci::estimate_mdiff_paired(
  comparison_mean = 14.25,
  comparison_sd = 4.28,
  reference_mean = 12.88,
  reference_sd = 3.4,
  n = 16,
  correlation = 0.87072223749,
  comparison_measure_name = "After",
  reference_measure_name = "Before"
)

# To visualize the estimated mean difference
myplot_from_summary <- esci::plot_mdiff(
  estimate_from_summary,
  effect_size = "mean"
)

# To conduct a hypothesis test
res_htest_from_summary <- esci::test_mdiff(
  estimate_from_summary,
  effect_size = "mean",
  rope = c(-2, 2)
)

Estimates for a two-group study with a continuous outcome variable

Description

Returns object estimate_mdiff_two is suitable for a simple two-group design with a continuous outcome variable. It provides estimates and CIs for the population mean difference between the repeated measures, the standardized mean difference (SMD; Cohen's d) between the repeated measures, and the median difference between the repeated measures (raw data only). You can pass raw data or summary data.

Usage

estimate_mdiff_two(
  data = NULL,
  outcome_variable = NULL,
  grouping_variable = NULL,
  comparison_mean = NULL,
  comparison_sd = NULL,
  comparison_n = NULL,
  reference_mean = NULL,
  reference_sd = NULL,
  reference_n = NULL,
  grouping_variable_levels = NULL,
  outcome_variable_name = "My outcome variable",
  grouping_variable_name = "My grouping variable",
  conf_level = 0.95,
  assume_equal_variance = FALSE,
  save_raw_data = TRUE,
  switch_comparison_order = FALSE
)

Arguments

Details

Reach for this function in place of an independent-samples t-test.

Once you generate an estimate with this function, you can visualize it with plot_mdiff() and you can test hypotheses with test_mdiff().

The estimated mean differences are from statpsych::ci.mean2().

The estimated SMDs are from CI_smd_ind_contrast().

The estimated median differences are from statpsych::ci.median2().

Value

Returns object of class esci_estimate

• es_mean_difference

  • type -

  • outcome_variable_name -

  • grouping_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • df -

  • ta_LL -

  • ta_UL -

• es_median_difference

  • type -

  • outcome_variable_name -

  • grouping_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • ta_LL -

  • ta_UL -

• es_smd

  • outcome_variable_name -

  • grouping_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • numerator -

  • denominator -

  • SE -

  • df -

  • d_biased -

• es_mean_ratio

  • outcome_variable_name -

  • grouping_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • comparison_mean -

  • reference_mean -

• es_median_ratio

  • outcome_variable_name -

  • grouping_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • comparison_median -

  • reference_median -

• overview

  • outcome_variable_name -

  • grouping_variable_name -

  • grouping_variable_level -

  • mean -

  • mean_LL -

  • mean_UL -

  • median -

  • median_LL -

  • median_UL -

  • sd -

  • min -

  • max -

  • q1 -

  • q3 -

  • n -

  • missing -

  • df -

  • mean_SE -

  • median_SE -

• raw_data

  • grouping_variable -

  • outcome_variable -

Examples

# From raw data
data("data_penlaptop1")

estimate_from_raw <- esci::estimate_mdiff_two(
  data = data_penlaptop1,
  outcome_variable = transcription,
  grouping_variable = condition,
  switch_comparison_order = TRUE,
  assume_equal_variance = TRUE
)

# To visualize the estimated median difference (raw data only)
myplot_from_raw <- esci::plot_mdiff(
  estimate_from_raw,
  effect_size = "median"
)

# To conduct a hypothesis test
res_htest_from_raw <- esci::test_mdiff(
  estimate_from_raw,
  effect_size = "median",
  rope = c(-2, 2)
)


# From summary data
estimate_from_summary <- esci::estimate_mdiff_two(
  comparison_mean = 12.09,
  comparison_sd = 5.52,
  comparison_n = 103,
  reference_mean = 6.88,
  reference_sd = 4.22,
  reference_n = 48,
  grouping_variable_levels = c("Ref-Laptop", "Comp-Pen"),
  outcome_variable_name = "% Transcription",
  grouping_variable_name = "Note-taking type",
  assume_equal_variance = TRUE
)

# To visualize the estimated mean difference
myplot <- esci::plot_mdiff(
  estimate_from_summary,
  effect_size = "mean"
)

# To conduct a hypothesis test
res_htest_from_summary <- esci::test_mdiff(
  estimate_from_summary,
  effect_size = "mean",
  rope = c(-2, 2)
)

Estimates for a multi-group study with a categorical outcome variable

Description

Returns object estimate_pdiff_ind_contrast is suitable for a multi-group design (between subjects) with a categorical outcome variable. It accepts a user-defined set of contrast weights that allows estimation of any 1-df contrast. It can express estimates as a difference in proportions and as an odds ratio (2-group designs only). You can pass raw data or summary data.

Usage

estimate_pdiff_ind_contrast(
  data = NULL,
  outcome_variable = NULL,
  grouping_variable = NULL,
  cases = NULL,
  ns = NULL,
  contrast = NULL,
  case_label = 1,
  grouping_variable_levels = NULL,
  outcome_variable_name = "My outcome variable",
  grouping_variable_name = "My grouping variable",
  conf_level = 0.95,
  count_NA = FALSE
)

Arguments

Details

Once you generate an estimate with this function, you can visualize it with plot_mdiff() and you can test hypotheses with test_mdiff().

The estimated proportion differences are from statpsych::ci.lc.prop.bs().

The estimated odds ratios (if returned) are from statpsych::ci.oddsratio().

Value

Returns object of class esci_estimate

• es_proportion_difference

  • type -

  • outcome_variable_name -

  • case_label -

  • grouping_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • effect_size_adjusted -

  • ta_LL -

  • ta_UL -

• es_odds_ratio

  • outcome_variable_name -

  • case_label -

  • grouping_variable_name -

  • effect -

  • effect_size -

  • SE -

  • LL -

  • UL -

  • ta_LL -

  • ta_UL -

• overview

  • grouping_variable_name -

  • grouping_variable_level -

  • outcome_variable_name -

  • outcome_variable_level -

  • cases -

  • n -

  • P -

  • P_LL -

  • P_UL -

  • P_SE -

  • P_adjusted -

  • ta_LL -

  • ta_UL -

• es_phi

  • grouping_variable_name -

  • outcome_variable_name -

  • effect -

  • effect_size -

  • SE -

  • LL -

  • UL -

Examples

# From raw data
data("data_campus_involvement")

estimate_from_raw <- esci::estimate_pdiff_ind_contrast(
  esci::data_campus_involvement,
  CommuterStatus,
  Gender,
  contrast = c("Male" = -1, "Female" = 1)
)

# To visualize the estimate
myplot_from_raw <- esci::plot_pdiff(estimate_from_raw)

# To conduct a hypothesis test
res_htest_from_raw <- esci::test_pdiff(estimate_from_raw)


# From summary data
estimate_from_summary <- esci::estimate_pdiff_ind_contrast(
  cases = c(78, 10),
  ns = c(252, 20),
  case_label = "egocentric",
  grouping_variable_levels = c("Original", "Replication"),
  contrast = c(-1, 1),
  conf_level = 0.95
)

# To visualize the estimate
myplot_from_summary <- esci::plot_pdiff(estimate_from_summary)

# To conduct a hypothesis test
res_htest_from_summary <- esci::test_pdiff(estimate_from_summary)

Estimates for a single-group design with a categorical outcome variable compared to a reference or population value.

Description

Returns object estimate_pdiff_one is suitable for a single-group design (between subjects) with a categorical outcome variable. It calculates the effect sizes with respect to a reference or population proportion (default value of 0). It returns the estimated difference between the in proportion from the reference/population value. You can pass raw data or summary data.

Usage

estimate_pdiff_one(
  data = NULL,
  outcome_variable = NULL,
  comparison_cases = NULL,
  comparison_n = NULL,
  reference_p = 0,
  case_label = 1,
  outcome_variable_name = "My outcome variable",
  conf_level = 0.95,
  count_NA = FALSE
)

Arguments

Details

Once you generate an estimate with this function, you can visualize it with plot_pdiff() and you can test hypotheses with test_pdiff().

The estimated proportion differences are from statpsych::ci.prop1() (renamed ci.prop as of statpsych 1.6).

Value

Returns an object of class esci_estimate

• overview

  • outcome_variable_name -

  • outcome_variable_level -

  • cases -

  • n -

  • P -

  • P_LL -

  • P_UL -

  • P_SE -

  • P_adjusted -

  • ta_LL -

  • ta_UL -

• es_proportion_difference

  • outcome_variable_name -

  • case_label -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • effect_size_adjusted -

  • ta_LL -

  • ta_UL -

  • cases -

  • n -

  • type -

Examples

# From raw data
data("data_campus_involvement")

estimate_from_raw <- esci::estimate_pdiff_one(
  esci::data_campus_involvement,
  CommuterStatus,
  reference_p = 0.50
)

# To visualize the estimate
myplot_from_raw <- esci::plot_pdiff(estimate_from_raw)

# To conduct a hypothesis test
res_htest_from_raw <- esci::test_pdiff(estimate_from_raw)


# From summary data
estimate_from_summary  <- esci::estimate_pdiff_one(
  comparison_cases = 8,
  comparison_n = 22,
  reference_p = 0.5
)

# To visualize the estimate
myplot_from_summary <- esci::plot_pdiff(estimate_from_summary)

# To conduct a hypothesis test
res_htest_from_summary <- esci::test_pdiff(estimate_from_summary)

Estimates for a repeated-measures study with two measures of a categorical variable

Description

Returns object estimate_pdiff_paired is suitable for a simple paired design with a categorical outcome variable. It provides estimates and CIs for the population proportion difference between the repeated measures. You can pass raw data or summary data.

Usage

estimate_pdiff_paired(
  data = NULL,
  comparison_measure = NULL,
  reference_measure = NULL,
  cases_consistent = NULL,
  cases_inconsistent = NULL,
  not_cases_consistent = NULL,
  not_cases_inconsistent = NULL,
  case_label = 1,
  not_case_label = NULL,
  comparison_measure_name = "Comparison measure",
  reference_measure_name = "Reference measure",
  conf_level = 0.95,
  count_NA = FALSE
)

Arguments

Details

Once you generate an estimate with this function, you can visualize it with plot_pdiff() and you can test hypotheses with test_pdiff().

The estimated proportion differences are from statpsych::ci.prop.ps().

Value

Returns object of class esci_estimate

Examples

# From summary data
# Example 1 from Bonett & Price, 2012
estimate_from_summary <- esci::estimate_pdiff_paired(
  cases_consistent = 60,
  cases_inconsistent = 50,
  not_cases_inconsistent = 22,
  not_cases_consistent = 68,
  case_label = "Answered True",
  not_case_label = "Answered False",
  reference_measure_name = "9th grade",
  comparison_measure_name = "12th grade",
  conf_level = 0.95
)

# To visualize the estimate
myplot_from_summary <- esci::plot_pdiff(estimate_from_summary)

# To conduct a hypothesis test
res_htest_from_summary <- esci::test_pdiff(estimate_from_summary)

Estimates for a two-group study with a categorical outcome variable

Description

Returns object estimate_pdiff_two is suitable for a simple two-group design with a categorical outcome variable. It provides estimates and CIs for the difference in proportions between the two groups, the odds ratio, and phi. You can pass raw data or summary data.

Usage

estimate_pdiff_two(
  data = NULL,
  outcome_variable = NULL,
  grouping_variable = NULL,
  comparison_cases = NULL,
  comparison_n = NULL,
  reference_cases = NULL,
  reference_n = NULL,
  case_label = 1,
  not_case_label = NULL,
  grouping_variable_levels = NULL,
  outcome_variable_name = "My outcome variable",
  grouping_variable_name = "My grouping variable",
  conf_level = 0.95,
  count_NA = FALSE
)

Arguments

Details

Once you generate an estimate with this function, you can visualize it with plot_mdiff() and you can test hypotheses with test_mdiff().

The estimated mean differences are from statpsych::ci.prop2().

The estimated odds ratio is from statpsych::ci.oddsratio().

The estimated correlation (phi) is from statpsych::ci.phi().

Value

Returns object of class esci_estimate

• es_proportion_difference

  • type -

  • outcome_variable_name -

  • case_label -

  • grouping_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • effect_size_adjusted -

  • ta_LL -

  • ta_UL -

• es_odds_ratio

  • outcome_variable_name -

  • case_label -

  • grouping_variable_name -

  • effect -

  • effect_size -

  • SE -

  • LL -

  • UL -

  • ta_LL -

  • ta_UL -

• overview

  • grouping_variable_name -

  • grouping_variable_level -

  • outcome_variable_name -

  • outcome_variable_level -

  • cases -

  • n -

  • P -

  • P_LL -

  • P_UL -

  • P_SE -

  • P_adjusted -

  • ta_LL -

  • ta_UL -

• es_phi

  • grouping_variable_name -

  • outcome_variable_name -

  • effect -

  • effect_size -

  • SE -

  • LL -

  • UL -

Examples

data("data_campus_involvement")

estimate_from_raw <- esci::estimate_pdiff_two(
  esci::data_campus_involvement,
  CommuterStatus,
  Gender
)

# To visualize the estimate
myplot_from_raw <- esci::plot_pdiff(estimate_from_raw)

# To conduct a hypothesis test
res_htest_from_raw <- esci::test_pdiff(estimate_from_raw)


# From summary_data
estimate_from_summary <- esci::estimate_pdiff_two(
  comparison_cases = 10,
  comparison_n = 20,
  reference_cases = 78,
  reference_n = 252,
  grouping_variable_levels = c("Original", "Replication"),
  conf_level = 0.95
)

# To visualize the estimate
myplot_from_summary <- esci::plot_pdiff(estimate_from_summary)

#' # To conduct a hypothesis test
res_htest_from_summary <- esci::test_pdiff(estimate_from_summary)

Estimates for a categorical variable with no grouping (single-group design)

Description

estimate_proportion is suitable for a single group design with a categorical outcome variable. It estimates the population proportion for the frequency of each level of the outcome variable, with confidence intervals. You can pass raw data or summary data.

Usage

estimate_proportion(
  data = NULL,
  outcome_variable = NULL,
  cases = NULL,
  case_label = 1,
  outcome_variable_levels = NULL,
  outcome_variable_name = "My outcome variable",
  conf_level = 0.95,
  count_NA = FALSE
)

Arguments

Details

Once you generate an estimate with this function, you can visualize it with plot_proportion().

If you want to compare your estimate to a known value or reference, then use estimate_pdiff_one().

The estimated proportions are from statpsych::ci.prop1() (renamed ci.prop as of statpsych 1.6).

Value

Returns an object of class esci_estimate

• overview

  • outcome_variable_name -

  • outcome_variable_level -

  • cases -

  • n -

  • P -

  • P_LL -

  • P_UL -

  • P_SE -

  • P_adjusted -

  • ta_LL -

  • ta_UL -

• es_proportion

  • outcome_variable_name -

  • case_label -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • effect_size_adjusted -

  • ta_LL -

  • ta_UL -

  • cases -

  • n -

Examples

# From raw data
data("data_campus_involvement")

estimate_from_raw <- esci::estimate_proportion(
  esci::data_campus_involvement,
  CommuterStatus
)


# To visualize the estimate
myplot_from_raw <- esci::plot_proportion(estimate_from_raw)


# From summary data
estimate_from_summary <- esci::estimate_proportion(
  cases = c(8, 22-8),
  outcome_variable_levels = c("Affected", "Not Affected")
)

# To visualize the estimate
myplot_from_summary<- esci::plot_proportion(estimate_from_summary)

Estimates the linear correlation (Pearson's r) between two continuous variables

Description

estimate_r is suitable for a design with two continuous variables. It estimates the linear correlation between two variables (Pearson's r) with a confidence interval. You can pass raw data or summary data.

Usage

estimate_r(
  data = NULL,
  x = NULL,
  y = NULL,
  r = NULL,
  n = NULL,
  x_variable_name = "My x variable",
  y_variable_name = "My y variable",
  conf_level = 0.95,
  save_raw_data = TRUE
)

Arguments

Details

Reach for this function to conduct simple linear correlation or simple linear regression.

Once you generate an estimate with this function, you can visualize it with plot_correlation() and you can test hypotheses with test_correlation(). In addition, you can use plot_scatter() to visualize the raw data and to conduct a regression analysis that r returns predicted Y' values from a given X value.

The estimated correlation is from statpsych::ci.cor(), which uses the Fisher r-to-z approach.

Value

Returns object of class esci_estimate

• overview

  • outcome_variable_name -

  • mean -

  • mean_LL -

  • mean_UL -

  • median -

  • median_LL -

  • median_UL -

  • sd -

  • min -

  • max -

  • q1 -

  • q3 -

  • n -

  • missing -

  • df -

  • mean_SE -

  • median_SE -

• es_r

  • x_variable_name -

  • y_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • n -

  • df -

  • ta_LL -

  • ta_UL -

• regression

  • component -

  • values -

  • LL -

  • UL -

• raw_data

  • x -

  • y -

  • fit -

  • lwr -

  • upr -

Examples

# From raw data
data("data_thomason_1")

estimate_from_raw <- esci::estimate_r(
  esci::data_thomason_1,
  Pretest,
  Posttest
)

# To visualize the value of r
myplot_correlation <- esci::plot_correlation(estimate_from_raw)

# To visualize the data (scatterplot) and use regression to obtain Y' from X
myplot_scatter_from_raw <- esci::plot_scatter(estimate_from_raw, predict_from_x = 10)

# To evaluate a hypothesis (interval null from -0.1 to 0.1):
res_htest_from_raw <- esci::test_correlation(
  estimate_from_raw,
  rope = c(-0.1, 0.1)
)


# From summary data
estimate_from_summary <- esci::estimate_r(r = 0.536, n = 50)

# To visualize the value of r
myplot_correlation_from_summary <- esci::plot_correlation(estimate_from_summary)

# To evaluate a hypothesis (interval null from -0.1 to 0.1):
res_htest_from_summary <- esci::test_correlation(
  estimate_from_summary,
  rope = c(-0.1, 0.1)
)

Estimates the difference in correlation for a design with two groups and two continuous outcome variables

Description

Returns object estimate_rdiff_two is suitable for a simple two-group design with two continuous outcome variables where you want to estimate the difference in the strength of the relationship between the two groups. It estimate the linear correlation (Pearson's r) for each group and the difference in r, along with confidence intervals. You can pass raw data or summary data.

Returns effect sizes appropriate for estimating the linear relationship between two quantitative variables

Usage

estimate_rdiff_two(
  data = NULL,
  x = NULL,
  y = NULL,
  grouping_variable = NULL,
  comparison_r = NULL,
  comparison_n = NULL,
  reference_r = NULL,
  reference_n = NULL,
  grouping_variable_levels = NULL,
  x_variable_name = "My x variable",
  y_variable_name = "My y variable",
  grouping_variable_name = "My grouping variable",
  conf_level = 0.95,
  save_raw_data = TRUE
)

Arguments

Details

Once you generate an estimate with this function, you can visualize it with plot_rdiff() and you can test hypotheses with test_rdiff(). In addition, you can use plot_scatter() to visualize the raw data.

The estimated single-group r values are from statpsych::ci.cor().

The difference in r values is from statpsych::ci.cor2().

Value

Returns object of class esci_estimate

• overview

  • outcome_variable_name -

  • grouping_variable_name -

  • grouping_variable_level -

  • mean -

  • mean_LL -

  • mean_UL -

  • median -

  • median_LL -

  • median_UL -

  • sd -

  • min -

  • max -

  • q1 -

  • q3 -

  • n -

  • missing -

  • df -

  • mean_SE -

  • median_SE -

• es_r_difference

  • type -

  • grouping_variable_name -

  • grouping_variable_level -

  • x_variable_name -

  • y_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • n -

  • df -

  • ta_LL -

  • ta_UL -

  • rz -

  • sem -

  • z -

  • p -

• es_r

  • grouping_variable_name -

  • grouping_variable_level -

  • x_variable_name -

  • y_variable_name -

  • effect -

  • effect_size -

  • LL -

  • UL -

  • SE -

  • n -

  • df -

  • ta_LL -

  • ta_UL -

• raw_data

  • x -

  • y -

  • grouping_variable -

Examples

# From raw data
data("data_campus_involvement")

estimate_from_raw <- esci::estimate_rdiff_two(
  esci::data_campus_involvement,
  GPA,
  SWB,
  Gender
)

# To visualize the difference in r
myplot_from_raw <- esci::plot_rdiff(estimate_from_raw)

# To visualize the data (scatterplot) by group
myplot_scatter <- esci::plot_scatter(estimate_from_raw)

# To evaluate a hypothesis (by default: point null of exaclty 0):
res_htest_from_raw <- esci::test_rdiff(
  estimate_from_raw
)


# From summary data
estimate <- esci::estimate_rdiff_two(
  comparison_r = .53,
  comparison_n = 45,
  reference_r = .41,
  reference_n = 59,
  grouping_variable_levels = c("Females", "Males"),
  x_variable_name = "Satisfaction with life",
  y_variable_name = "Body satisfaction",
  grouping_variable_name = "Gender",
  conf_level = .95
)

myplot_from_summary <- esci::plot_rdiff(estimate)

# To evaluate a hypothesis (interval null from -0.1 to 0.1):
res_htest_from_summary <- esci::test_rdiff(
  estimate,
  rope = c(-0.1, 0.1)
)

Meta-analysis diamond

Description

geom_meta_diamond_h creates a horizontal meta-analytic diamond defined by an x value (horizontal center of diamond), xmin and xmax values (for the horizontal ends of the diamond), a y value (for the vertical placement of the diamond) and a height (for the vertical height of the diamond). Note the use of xmin and xmax allows for representation of asymmetric confidence intervals with this geom.

Usage

geom_meta_diamond_h(
  mapping = NULL,
  data = NULL,
  stat = "identity",
  position = "identity",
  ...,
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = TRUE
)

Arguments

Examples

# example code
library(ggplot2)

my_effects <- data.frame(
  effect_size = c(1, 2, 1, 0),
  UL = c(2, 3, 2, 1),
  LL = c(0, 1, 0, -1),
  y = c(1, 2, 3, 4)
)


myplot <- ggplot2::ggplot()
myplot <- myplot + geom_meta_diamond_h(
  data = my_effects,
  ggplot2::aes(
    x = effect_size,
    xmin = LL,
    xmax = UL,
    y = y
  ),
  height = 0.25,
  color = "black",
  fill = "red",
)

Correlations: Single Group

Description

Correlations: Single Group

Usage

jamovicorrelation(
  switch = "from_raw",
  data,
  x,
  y,
  r = " ",
  n = " ",
  x_variable_name = "X variable",
  y_variable_name = "Y variable",
  conf_level = 95,
  show_details = FALSE,
  do_regression = FALSE,
  show_line = FALSE,
  show_line_CI = FALSE,
  show_residuals = FALSE,
  show_PI = FALSE,
  show_mean_lines = FALSE,
  show_r = FALSE,
  plot_as_z = FALSE,
  predict_from_x = " ",
  evaluate_hypotheses = FALSE,
  null_value = "0",
  null_boundary = "0",
  alpha = 0.05,
  null_color = "#A40122",
  es_plot_width = "300",
  es_plot_height = "400",
  sp_plot_width = "650",
  sp_plot_height = "650",
  ymin = "-1",
  ymax = "1",
  ybreaks = "auto",
  ylab = "auto",
  xlab = "auto",
  axis.text.y = "14",
  axis.title.y = "15",
  axis.text.x = "14",
  axis.title.x = "15",
  error_layout = "none",
  sp_ymin = "auto",
  sp_ymax = "auto",
  sp_ybreaks = "auto",
  sp_xmin = "auto",
  sp_xmax = "auto",
  sp_xbreaks = "auto",
  sp_ylab = "auto",
  sp_xlab = "auto",
  sp_axis.text.y = "14",
  sp_axis.title.y = "15",
  sp_axis.text.x = "14",
  sp_axis.title.x = "15",
  shape_summary = "circle filled",
  color_summary = "#008DF9",
  fill_summary = "#008DF9",
  size_summary = "4",
  alpha_summary = "1",
  linetype_summary = "solid",
  color_interval = "black",
  size_interval = "3",
  alpha_interval = "1",
  alpha_error = "1",
  fill_error = "gray75",
  sp_shape_raw_reference = "circle filled",
  sp_color_raw_reference = "black",
  sp_fill_raw_reference = "#008DF9",
  sp_size_raw_reference = "3",
  sp_alpha_raw_reference = ".25",
  sp_linetype_summary_reference = "solid",
  sp_color_summary_reference = "#008DF9",
  sp_size_summary_reference = "3",
  sp_alpha_summary_reference = ".25",
  sp_linetype_PI_reference = "dotted",
  sp_color_PI_reference = "#E20134",
  sp_size_PI_reference = "2",
  sp_alpha_PI_reference = "1",
  sp_linetype_residual_reference = "solid",
  sp_color_residual_reference = "#E20134",
  sp_size_residual_reference = "1",
  sp_alpha_residual_reference = "1",
  sp_prediction_label = "5",
  sp_prediction_color = "#E20134",
  sp_linetype_CI = "solid",
  sp_color_CI = "#008DF9",
  sp_size_CI = "4",
  sp_alpha_CI = "1",
  sp_linetype_ref = "dotted",
  sp_color_ref = "gray60",
  sp_size_ref = "1",
  sp_alpha_ref = "1",
  sp_linetype_PI = "solid",
  sp_color_PI = "#E20134",
  sp_size_PI = "2",
  sp_alpha_PI = "1"
)

Arguments

Value

A results object containing:

Tables can be converted to data frames with asDF or as.data.frame. For example:

results$overview$asDF

as.data.frame(results$overview)

Describe

Description

Describe

Usage

jamovidescribe(
  data,
  outcome_variable,
  show_details = FALSE,
  mark_mean = FALSE,
  mark_median = FALSE,
  mark_sd = FALSE,
  mark_quartiles = FALSE,
  mark_z_lines = FALSE,
  mark_percentile = "0",
  histogram_bins = "12",
  es_plot_width = "500",
  es_plot_height = "400",
  ymin = "auto",
  ymax = "auto",
  breaks = "auto",
  xmin = "auto",
  xmax = "auto",
  xbreaks = "auto",
  ylab = "auto",
  xlab = "auto",
  axis.text.y = "14",
  axis.title.y = "15",
  axis.text.x = "14",
  axis.title.x = "15",
  fill_regular = "#008DF9",
  fill_highlighted = "#E20134",
  color = "black"
)

Arguments

Value

A results object containing:

Tables can be converted to data frames with asDF or as.data.frame. For example:

results$overview$asDF

as.data.frame(results$overview)

Means and Medians: Single Group

Description

Means and Medians: Single Group

Usage

jamovimagnitude(
  switch = "from_raw",
  data,
  outcome_variable,
  mean = "",
  sd = "",
  n = "",
  outcome_variable_name = "Outcome variable",
  conf_level = 95,
  effect_size = "mean",
  show_details = FALSE,
  show_calculations = FALSE,
  evaluate_hypotheses = FALSE,
  null_value = "0",
  null_boundary = "0",
  alpha = 0.05,
  null_color = "#A40122",
  es_plot_width = "300",
  es_plot_height = "400",
  ymin = "auto",
  ymax = "auto",
  breaks = "auto",
  ylab = "auto",
  xlab = "auto",
  axis.text.y = "14",
  axis.title.y = "15",
  axis.text.x = "14",
  axis.title.x = "15",
  error_layout = "halfeye",
  error_scale = "0.20",
  error_nudge = "0.3",
  data_layout = "random",
  data_spread = "0.25",
  shape_raw = "circle filled",
  shape_summary = "circle filled",
  color_raw = "#008DF9",
  color_summary = "#008DF9",
  fill_raw = "NA",
  fill_summary = "#008DF9",
  size_raw = "2",
  size_summary = "4",
  alpha_raw = "1",
  alpha_summary = "1",
  linetype_summary = "solid",
  color_interval = "black",
  size_interval = "3",
  alpha_interval = "1",
  alpha_error = "1",
  fill_error = "gray75"
)

Arguments

Value

A results object containing:

Tables can be converted to data frames with asDF or as.data.frame. For example:

results$overview$asDF

as.data.frame(results$overview)

Means and Medians: 2x2 Factorial

Description

Means and Medians: 2x2 Factorial

Usage

jamovimdiff2x2(
  design = "fully_between",
  switch = "from_raw",
  data,
  outcome_variable,
  grouping_variable_A,
  grouping_variable_B,
  outcome_variable_level1,
  outcome_variable_level2,
  outcome_variable_name_bs = "My outcome variable",
  grouping_variable,
  repeated_measures_name = "Time",
  outcome_variable_name = "My outcome variable",
  A1_label = "A1 level",
  A2_label = "A2 level",
  B1_label = "B1 level",
  B2_label = "B2 level",
  A_label = "Variable A",
  B_label = "Variable B",
  A1B1_mean = " ",
  A1B1_sd = " ",
  A1B1_n = " ",
  A1B2_mean = " ",
  A1B2_sd = " ",
  A1B2_n = " ",
  A2B1_mean = " ",
  A2B1_sd = " ",
  A2B1_n = " ",
  A2B2_mean = " ",
  A2B2_sd = " ",
  A2B2_n = " ",
  conf_level = 95,
  effect_size = "mean_difference",
  assume_equal_variance = TRUE,
  show_details = FALSE,
  show_interaction_plot = FALSE,
  show_CI = FALSE,
  evaluate_hypotheses = FALSE,
  null_value = "0",
  null_boundary = "0",
  rope_units = "raw",
  alpha = 0.05,
  null_color = "#A40122",
  es_plot_width = "700",
  es_plot_height = "400",
  ymin = "auto",
  ymax = "auto",
  ybreaks = "auto",
  ylab = "auto",
  xlab = "auto",
  axis.text.y = "14",
  axis.title.y = "15",
  axis.text.x = "14",
  axis.title.x = "15",
  simple_contrast_labels = FALSE,
  error_layout = "halfeye",
  error_scale = "0.25",
  error_nudge = "0.5",
  data_layout = "random",
  data_spread = "0.20",
  difference_axis_units = "raw",
  difference_axis_breaks = "auto",
  shape_raw_reference = "circle filled",
  shape_raw_comparison = "circle filled",
  shape_raw_unused = "circle filled",
  shape_summary_reference = "circle filled",
  shape_summary_comparison = "circle filled",
  shape_summary_unused = "circle filled",
  shape_summary_difference = "triangle filled",
  color_raw_reference = "#008DF9",
  color_raw_comparison = "#009F81",
  color_raw_unused = "gray65",
  color_summary_reference = "#008DF9",
  color_summary_comparison = "#009F81",
  color_summary_unused = "gray65",
  color_summary_difference = "black",
  fill_raw_reference = "NA",
  fill_raw_comparison = "NA",
  fill_raw_unused = "NA",
  fill_summary_reference = "#008DF9",
  fill_summary_comparison = "#009F81",
  fill_summary_unused = "gray65",
  fill_summary_difference = "black",
  size_raw_reference = "2",
  size_raw_comparison = "2",
  size_raw_unused = "1",
  size_summary_reference = "4",
  size_summary_comparison = "4",
  size_summary_unused = "4",
  size_summary_difference = "4",
  alpha_raw_reference = "1",
  alpha_raw_comparison = "1",
  alpha_raw_unused = "1",
  alpha_summary_reference = "1",
  alpha_summary_comparison = "1",
  alpha_summary_unused = "1",
  alpha_summary_difference = "1",
  linetype_summary_reference = "solid",
  linetype_summary_comparison = "solid",
  linetype_summary_unused = "solid",
  linetype_summary_difference = "solid",
  color_interval_reference = "black",
  color_interval_comparison = "black",
  color_interval_unused = "gray65",
  color_interval_difference = "black",
  size_interval_reference = "3",
  size_interval_comparison = "3",
  size_interval_unused = "3",
  size_interval_difference = "3",
  alpha_interval_reference = "1",
  alpha_interval_comparison = "1",
  alpha_interval_unused = "1",
  alpha_interval_difference = "1",
  alpha_error_reference = "1",
  alpha_error_comparison = "1",
  alpha_error_unused = "1",
  alpha_error_difference = "1",
  fill_error_reference = "gray75",
  fill_error_comparison = "gray75",
  fill_error_unused = "gray75",
  fill_error_difference = "gray75"
)

Arguments

Value

A results object containing:

Tables can be converted to data frames with asDF or as.data.frame. For example:

results$overview$asDF

as.data.frame(results$overview)

Means and Medians: Independent Groups Contrast

Description

Means and Medians: Independent Groups Contrast

Usage

jamovimdiffindcontrast(
  switch = "from_raw",
  data,
  outcome_variable,
  grouping_variable,
  means,
  sds,
  ns,
  grouping_variable_levels,
  outcome_variable_name = "Outcome variable",
  grouping_variable_name = "Grouping variable",
  comparison_labels = " ",
  reference_labels = " ",
  conf_level = 95,
  effect_size = "mean_difference",
  assume_equal_variance = TRUE,
  show_details = FALSE,
  evaluate_hypotheses = FALSE,
  null_value = "0",
  null_boundary = "0",
  rope_units = "raw",
  alpha = 0.05,
  null_color = "#A40122",
  es_plot_width = "550",
  es_plot_height = "450",
  ymin = "auto",
  ymax = "auto",
  ybreaks = "auto",
  ylab = "auto",
  xlab = "auto",
  axis.text.y = "14",
  axis.title.y = "15",
  axis.text.x = "14",
  axis.title.x = "15",
  simple_contrast_labels = TRUE,
  error_layout = "halfeye",
  error_scale = "0.25",
  error_nudge = "0.5",
  data_layout = "random",
  data_spread = "0.20",
  difference_axis_units = "raw",
  difference_axis_breaks = "auto",
  shape_raw_reference = "circle filled",
  shape_raw_comparison = "circle filled",
  shape_raw_unused = "circle filled",
  shape_summary_reference = "circle filled",
  shape_summary_comparison = "circle filled",
  shape_summary_unused = "circle filled",
  shape_summary_difference = "triangle filled",
  color_raw_reference = "#008DF9",
  color_raw_comparison = "#009F81",
  color_raw_unused = "gray65",
  color_summary_reference = "#008DF9",
  color_summary_comparison = "#009F81",
  color_summary_unused = "gray65",
  color_summary_difference = "black",
  fill_raw_reference = "NA",
  fill_raw_comparison = "NA",
  fill_raw_unused = "NA",
  fill_summary_reference = "#008DF9",
  fill_summary_comparison = "#009F81",
  fill_summary_unused = "gray65",
  fill_summary_difference = "black",
  size_raw_reference = "2",
  size_raw_comparison = "2",
  size_raw_unused = "1",
  size_summary_reference = "4",
  size_summary_comparison = "4",
  size_summary_unused = "4",
  size_summary_difference = "4",
  alpha_raw_reference = "1",
  alpha_raw_comparison = "1",
  alpha_raw_unused = "1",
  alpha_summary_reference = "1",
  alpha_summary_comparison = "1",
  alpha_summary_unused = "1",
  alpha_summary_difference = "1",
  linetype_summary_reference = "solid",
  linetype_summary_comparison = "solid",
  linetype_summary_unused = "solid",
  linetype_summary_difference = "solid",
  color_interval_reference = "black",
  color_interval_comparison = "black",
  color_interval_unused = "gray65",
  color_interval_difference = "black",
  size_interval_reference = "3",
  size_interval_comparison = "3",
  size_interval_unused = "3",
  size_interval_difference = "3",
  alpha_interval_reference = "1",
  alpha_interval_comparison = "1",
  alpha_interval_unused = "1",
  alpha_interval_difference = "1",
  alpha_error_reference = "1",
  alpha_error_comparison = "1",
  alpha_error_unused = "1",
  alpha_error_difference = "1",
  fill_error_reference = "gray75",
  fill_error_comparison = "gray75",
  fill_error_unused = "gray75",
  fill_error_difference = "gray75"
)

Arguments

Value

A results object containing:

Tables can be converted to data frames with asDF or as.data.frame. For example:

results$overview$asDF

as.data.frame(results$overview)

Means and Medians: Paired

Description

Means and Medians: Paired

Usage

jamovimdiffpaired(
  switch = "from_raw",
  data,
  reference_measure,
  comparison_measure,
  comparison_mean = " ",
  comparison_sd = " ",
  reference_mean = " ",
  reference_sd = " ",
  n = " ",
  enter_r_or_sdiff = "enter_r",
  correlation = " ",
  sdiff = " ",
  comparison_measure_name = "Comparison measure",
  reference_measure_name = "Reference measure",
  conf_level = 95,
  effect_size = "mean_difference",
  show_ratio = FALSE,
  show_details = FALSE,
  show_calculations = FALSE,
  evaluate_hypotheses = FALSE,
  null_value = "0",
  null_boundary = "0",
  rope_units = "raw",
  alpha = 0.05,
  null_color = "#A40122",
  es_plot_width = "600",
  es_plot_height = "400",
  ymin = "auto",
  ymax = "auto",
  ybreaks = "auto",
  ylab = "auto",
  xlab = "auto",
  axis.text.y = "14",
  axis.title.y = "15",
  axis.text.x = "14",
  axis.title.x = "15",
  simple_contrast_labels = TRUE,
  error_layout = "halfeye",
  error_scale = "0.25",
  error_nudge = "0.5",
  data_layout = "random",
  data_spread = "0.20",
  difference_axis_units = "raw",
  difference_axis_breaks = "auto",
  shape_raw_reference = "circle filled",
  shape_raw_comparison = "circle filled",
  shape_raw_difference = "triangle filled",
  shape_summary_reference = "circle filled",
  shape_summary_comparison = "circle filled",
  shape_summary_difference = "triangle filled",
  color_raw_reference = "#008DF9",
  color_raw_comparison = "#008DF9",
  color_raw_difference = "#E20134",
  color_summary_reference = "#008DF9",
  color_summary_comparison = "#008DF9",
  color_summary_difference = "black",
  fill_raw_reference = "NA",
  fill_raw_comparison = "NA",
  fill_raw_difference = "NA",
  fill_summary_reference = "#008DF9",
  fill_summary_comparison = "#008DF9",
  fill_summary_difference = "black",
  size_raw_reference = "2",
  size_raw_comparison = "2",
  size_raw_difference = "2",
  size_summary_reference = "4",
  size_summary_comparison = "4",
  size_summary_difference = "4",
  alpha_raw_reference = "1",
  alpha_raw_comparison = "1",
  alpha_raw_difference = "1",
  alpha_summary_reference = "1",
  alpha_summary_comparison = "1",
  alpha_summary_difference = "1",
  linetype_summary_reference = "solid",
  linetype_summary_comparison = "solid",
  linetype_summary_difference = "solid",
  color_interval_reference = "black",
  color_interval_comparison = "black",
  color_interval_difference = "black",
  size_interval_reference = "3",
  size_interval_comparison = "3",
  size_interval_difference = "3",
  alpha_interval_reference = "1",
  alpha_interval_comparison = "1",
  alpha_interval_difference = "1",
  alpha_error_reference = "1",
  alpha_error_comparison = "1",
  alpha_error_difference = "1",
  fill_error_reference = "gray75",
  fill_error_comparison = "gray75",
  fill_error_difference = "gray75"
)

Arguments