[R-meta] Interpreting multi-level eggers test

[Röhl, Sebastian]

Hi Katherine,

I would interpret it as follows:
Studies with larger SEs show a significantly larger effect size. Lager SEs mean smaller sample sizes.
Therefore, the test indicates a probable publication bias, as small studies with small (and therefore non-significant) effects are less present in your study pool.

Best,
Sebastian


****************************
Dr. Sebastian Röhl
Eberhard Karls Universität Tübingen
Institute for Educational Science
Tübingen School of Education (TüSE)
Wilhelmstraße 31 / Room 302
D-72074 Tübingen
Germany

Phone: +49 7071 29-75527
Fax: +49 7071 29-35309
Email: sebastian.roehl using uni-tuebingen.de<mailto:sebastian.roehl using uni-tuebingen.de>
Twitter: @sebastian_roehl  @ResTeacherEdu



Von: R-sig-meta-analysis <r-sig-meta-analysis-bounces using r-project.org> Im Auftrag von Katherine via R-sig-meta-analysis
Gesendet: Dienstag, 14. Februar 2023 02:58
An: r-sig-meta-analysis using r-project.org
Cc: Katherine <manbe003 using uw.edu>
Betreff: [R-meta] Interpreting multi-level eggers test

Hi all,


I am trying to conduct a 3-level metaanalysis using rma.mv. I've been exploring options for detecting selective reporting and was planning to use eggers test as explored in a previous email on this listerv (12/9/2020).

The instructions given by James involve including the standard error of the effect size as a moderator in rma.mv, and then applying cluster-robust standard errors using coef_test.

I conducted this analysis and the output was:
Coef. Estimate    SE       t-stat d.f. (Satt) p-val (Satt) Sig.
intrcpt    0.0657    0.175     0.375           25.2        0.711
mods   2.2255   0.826     2.693           13.5        0.018    *

I'm not sure how to interpret this. I assume the significant estimate for the mods indicates bias or selective reporting?

Does anyone have resources for how to interpret these numbers, and what to do next? I am hoping to find outliers and rerun analyses without outliers, to determine the robustness of my results.

This is new to me, so I greatly appreciate any help you can provide!!

Thanks for considering,
Katherine


**************************************************

Katherine Manbeck, M.S. | Pronouns: they/them/she/her
PhD Candidate, Clinical Psychology
University of Washington
Phone: 952.240.8947

The University of Washington stands on sluʔwiɫ, a village within the traditional homelands of the Coast Salish people. The dxʷdəwʔabš (Duwamish) people past and present are the rightful stewards of this land. White settlers perpetuated genocide<https://kingcounty.gov/council/news/2015/February/native_expulsion.aspx> against Indigenous inhabitants of this area, convinced Indigenous leaders to sign predatory treaties<https://goia.wa.gov/tribal-government/treaty-point-elliott-1855>, and refused to honor<https://www.duwamishtribe.org/treaty-of-point-elliott> the meager terms of the treaties, including refusing to establish a reservation for the Duwamish people. Despite the legacy of genocide and displacement of Indigenous peoples in the Seattle area, the Duwamish are still here<https://www.duwamishtribe.org/>, a testament to their strength and resilience.
All settlers (non-Indigenous people) benefitting from this land, but especially descendants of colonizers, must find ways to support Indigenous sovereignty. The Land Back Movement<https://landback.org/> and the Land Reparations and Indigenous Solidarity Toolkit<https://resourcegeneration.org/land-reparations-indigenous-solidarity-action-guide/> offer concrete reparations for the ongoing illegal occupation of Indigenous lands across the United States. At a minimum, I believe that all settlers in the Seattle area should pay monthly Real Rent<https://www.realrentduwamish.org/> to the Duwamish tribe.
************************************************

The above email may contain patient identifiable or confidential information. Because email is not secure, please be aware of associated risks of email transmission. If you are a patient, communicating to a UW Healthcare Provider via email implies your agreement to email communication. See http://uwmedicine.washington.edu/Global/Compliance/Pages/Risks-of-Using-Email.aspx





The information is intended for the individual named above. If you are not the intended recipient, any disclosure, copying, distribution or use of the contents of this information is prohibited. Please notify the sender by reply email, and then destroy all copies of the message and any attachments. See our Notice of Privacy Practices at http://uwmedicine.washington.edu/Global/Compliance/Pages/Notice-of-Privacy-Practices.aspx




	[[alternative HTML version deleted]]