https://goo.gl/YcZhwy

The Social Psychology Lab

holistic views
Sep 9, 2018 · 3 min read

Everyone who’s part of the Social Psychology Lab, in some way or another, has worked on a project involving social behaviour, interactions and/or positive emotions. There are currently 3 honours students, 1 past honours student (now research assistant), 1 past research intern and a few others who joined along the way (e.g. volunteering during their undergraduate course).

It was interesting to learn that the lab meetings (and associated slack channel messages) maintain a fairly high engagement even though some members are no longer directly involved with studies. Perhaps because social psychology attracts social butterflies? 🤭

https://goo.gl/8pKncC

In our upcoming lab meeting, we’ll be teaching each other about our investigations into Cronbach’s Alpha vs. McDonald’s Omega for measuring internal consistency (also called scale reliability). You might have been thinking “Why are you doing this?”. Well, dear reader, you may not know this but there has been emerging literature highlighting the limitations and issues with the widely-used Cronbach’s Alpha in determining the internal consistency of a test (e.g. life satisfaction). As a result, researchers — including my supervisor — have begun considering alternative approaches to measure internal consistency of their own studies. One popular alternative is McDonald’s Omega.

Now for those (like me) who need a refresher on what internal consistency means, it can be defined as “how closely related a set of items are as a group”. E.g. we ask participants to answer 5 different questions to measure their level of life satisfaction. How well does each question actually measure the same construct?

We were also tasked with finding out how to calculate Omega with RStudio — which gave me an opportunity to put our early coding workshop skills to great use… Especially on how to install packages properly! I’ve now learnt that Cronbach’s Alpha is highly flawed as a measure of internal consistency compared to that of Mcdonald’s Omega, due to the different underlying assumptions of each approach (such as tau-equivalence, normality, uncorrelated errors).


Project update

With lots of help from my supervisor, I’ve created a study design plan / framework for my ‘vicarious pride expression questionnaire’ and also started programming in Qualtrics. I found so many cool features available, but unfortunately I won’t be able to use them due to time and project restraints.

Example of some question types, Qualtrics

When my next update rolls around, I hope to have launched the questionnaire onto MTurk (Amazon’s crowd-sourcing platform) and be prepared to submit my final research proposal.


For future research interns

It is 1000% okay if you’re feeling a bit lost or insecure about your own understanding of the topic at this point of semester, because I feel exactly the same way right now too. Just keep in mind that this is a research internship, so keep up the great work! You can do it ♪

https://goo.gl/5VGJZb

Hoping we can all handle the next few busy weeks!

-Joanne

holistic views

Written by

a psyc3361 blog by three research interns in social psychology, behavioural neuroscience, and cognitive science

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade