Missing sections

This article shows a possible workaround to a common problem encountered when dealing with a standard Munich ChronoType Questionnaire (MCTQ) and \(\mu\)MCTQ data.

It’s a good idea to have the standard MCTQ questionnaire and the guidelines for the standard MCTQ variable computation open while reading this article/vignette. That way you can have a better understanding of the data objects we are going to deal with. You can download a copy of the MCTQ full standard version here. Click here to download a copy of the guidelines for the standard MCTQ variables.

Working around missing sections

Although the standard and micro versions of the MCTQ asks for respondents to complete the workdays and work-free days sections, even when they do not have a regular work schedule (wd = 0) or have a 7 day/week work schedule (wd = 7), some of them may still end skipping one of this parts of the questionnaire. In those cases, sd_week(), sloss_week(), le_week(), msf_sc(), sjl_rel(), sjl(), sjl_sc_rel(), and sjl_sc() will produce NA (Not Available) as output. That’s because those computations combine workdays and work-free days variables.

For those special standard and micro MCTQ cases, where one section is missing, a NA value is the correct output for the functions mentioned above when wd (number of workdays per week) are wd > 0 & wd < 7, but it may not be when wd == 0 or wd == 7. While some researchers may just invalidate these latter cases, we propose a different approach.

To illustrate this approach, consider the following.

If a respondent does not have a regular work schedule (wd == 0), only answered the work-free days section, and does not use an alarm clock on their free days (i.e., alarm_f == FALSE), it would be fair to assume that there’s no sleep correction (sc) to be made, therefore, their chronotype (msf_sc) must be equal to their midsleep on work-free days (msf).

Following this same line of thought, we can also say that:

Note that the chronotype computation follows a similar line of thought.

The opposite scenario, i.e., when the respondent works 7 days per week (wd == 7) and only answered the workdays section, can also have different outputs. sloss_week(), msf_sc(), sjl_rel(), sjl(), sjl_sc_rel(), and sjl_sc() should still produce a NA as output since there’s no way to know the real behavior of the respondent’s sleep-wake cycle. But, according to this reasoning, sd_week and le_week can have different outputs.

If you agree with this line of thought, we recommend creating dummy variables to identify those two situations and then change the values as mentioned. You can see this procedure in action with the data wrangling algorithms made to produce the fictional std_mctq dataset, provided by the mctq package.

Please note that this workaround is not mentioned or endorsed by the MCTQ authors. If you use it, you must mention this reasoning in your methods section.