A brief post! I have been busy over summer but also taking some rest. I just wanted to let you know that we have submitted two new preprints to PsyArxiv.
In the first preprint, we examine the consequences of arbitrary binning survey data. It is quite common that you’ll read X % agree with the statement. Usually, however, the underlying question is a Likert scale (e.g., five response options ranging from strongly disagree to strongly disagree). For the National Student Survey, a nationwide UK survey on student satisfaction, it is common to report X% students are satisfied. However, these summaries tend to ignore the midpoint. Those who “pick neither agree nor disagree” are implicitly grouped as dissatisfied in the above summary. So, we set out what happens if one would have counted the mid-point as satisfied. It turns out that for individual courses or universities, this type of grouping can matter quite a bit for their rank.
In the second preprint, we examined whether Instagram use and the type of Instagram use matters for well-being. What is interesting about our study is that we relied on a sample which was nationally representative for the UK in terms of age and gender. So what did we find… ? Not much. Even though we have a sizable sample we did not find much evidence for a negative impact of self-reported Instagram use and type of use on well-being. Note that this work was financed by Facebook Research, so please see the disclosure statement but note that they played no role in study design, reporting of findings, etc.
Fingers crossed that some journals will like and publish our preprints… .
Anyway: Have a read! (Or don’t: it is time to enjoy the last of summer, I guess).