Good evening amazing souls!
I have a study of 150 participants, and there are 55 different ages (multiple in some groups), what is the best way to describe the age breakdown in my paper? Is it by decade, generation, or is there a procedural that is standard for combining into groups?
I have 2 Likert scales, and want to see if age, gender, education are factors in if the scores are low or high, yet not sure which test to run in SPSS - suggestions? I did try the continuous table, yet with 55 ages it looks massive. I did try a T-Test (single). Any insight would be greatly accepted.
Thank you.
Jason
Hello Jason David Miller
I would describe it in the text as average +/- standard deviation and then say (range: x to y). If the distribution is important to the topic and/or analysis, then I would additionally show it as a bar graph/figure (in decades ie 20-29, 20-39, 40-49 and so on) and refer to it in the text.
For descriptive stats you can define age groups. The groups should be oriented on yor research, if this is possible (e.g. if things like puperty play a role, then one of these age groups should comprise the typical ages when people are in puperty). Without any meaningful ideas you can define equally sized groups. If you make a median split you get two groups (the 50% "younger" and the 50% "older" individuals). With a tertile split you get trhee groups (33% "younger", 33% "middle aged" and 33% older). You can get finer resolution with quartile- pentile - etc. splits. However, I thin that a median or tertile split is usually sufficient.
For inferential stats you should use a continuous covariate. You can use splines (i.e., generalized* additive models) to allow an unspecified "wiggly" relationship between age and the response.
*As your responses are Likert scores that have a lower and an upper limit, you may use a generalized model with logit- or probit-link.
Age can simply be analyzed descriptively. i hope you had grouped your age in the tool
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression