Hello Sid,

A bit out of order, but here are some thoughts on your three questions.

Q2: You have to have responses to the measures in order to estimate Cronbach's alpha (or any indicator of response consistency), so you were correct in trying to collect some data. Otherwise, since you note these are existing measures (and, presumably, others have compiled evidence concerning their technical quality), all you'd have been able to do would be to report what others have found (with their samples, which might or might not be comparable to yours).

Q1: Is 64 cases enough? For a pilot study, it's certainly enough to permit you to estimate internal consistency reliability. When you finish with the "full sample," it would be good practice to report score reliability estimates based on the larger batch as well.

Q3: Whether 107 cases is enough for multiple regression analysis depends on: (a) how many IVs (and DVs) you're evaluating in your model(s); (b) the smallest effect size (for example, how low an R-squared) you would consider noteworthy given the variables and the population you're investigating; and (c) how much statistical power you'd like your study to have to detect that effect size, if it exists in the population (when testing at your chosen alpha level, of course). Programs like the freely available G*Power program can help with this process (https://www.psychologie.hhu.de/arbeitsgruppen/allgemeine-psychologie-und-arbeitspsychologie/gpower.html)

Good luck with your work.

Hello Sid,

A bit out of order, but here are some thoughts on your three questions.

Q2: You have to have responses to the measures in order to estimate Cronbach's alpha (or any indicator of response consistency), so you were correct in trying to collect some data. Otherwise, since you note these are existing measures (and, presumably, others have compiled evidence concerning their technical quality), all you'd have been able to do would be to report what others have found (with their samples, which might or might not be comparable to yours).

Q1: Is 64 cases enough? For a pilot study, it's certainly enough to permit you to estimate internal consistency reliability. When you finish with the "full sample," it would be good practice to report score reliability estimates based on the larger batch as well.

Q3: Whether 107 cases is enough for multiple regression analysis depends on: (a) how many IVs (and DVs) you're evaluating in your model(s); (b) the smallest effect size (for example, how low an R-squared) you would consider noteworthy given the variables and the population you're investigating; and (c) how much statistical power you'd like your study to have to detect that effect size, if it exists in the population (when testing at your chosen alpha level, of course). Programs like the freely available G*Power program can help with this process (https://www.psychologie.hhu.de/arbeitsgruppen/allgemeine-psychologie-und-arbeitspsychologie/gpower.html)

Good luck with your work.

Dear Sid Al-kanwal ,

Fully supporting the response from David Morse , I would pay your attention also to two other things:

1) Representatives of your sample to the population you conclude about. How you get the respondents? What was your sampling technique? (See e.g. Article Sampling Methods in Clinical Research; an Educational Review

2) How much the questionnaires you use are close to the ones with established psychometric measures? The same language? The same format (paper-and-pencil or web-based)? The same layout? Similar sample of respondents? If it is really the same, you don't need to repeat internal consistency estimation.

Good luck!

Dear Sid,

David Morse and Michal S. Karbownik has answered your questions very satisfactorily.

Yours questionnaire would have psychometric properties from the developer and other people that have used it before. You can report that it is an already standardized instrument.

However, the 67 sample is enough to calculate C-alpha in case you wish to check if yours results will be similar to the former validation. Otherwise, you can calculate the C-alpha post hoc (when you are done with your sample collection).

For the sample size you can download G*power 3.1.0 software, it is free. Check compute sample size. Select test family, chose regression, you will see an interphase to input your desired power (in % eg 80,90,95), alpha level of error ( may be 0.05) etc. Click compute and you we see the required sample size. It is very easy to use. You can get back to me if you require more clarity on G*power.

Good luck.

You can calculate the true Bayesian confidence intervals of reliability in JASP access free program, I also recommend using omega coefficient

Thank you so much David Morse Ogochukwu Kelechi Onyeso Michal S. Karbownik

Your help is greatly appreciated!

Another question, for multiple regression is a individually calculated bivariate correlation between each predictor and criterion required/beneficial?

Hello again Sid,

By way of descriptive statistics to be reported accompanying a linear regression analysis, it's always a good idea to report (at a minimum) means, SDs, and inter-correlations for all of the variables. So, yes to your follow-up question.

The individual IV-DV correlations can be helpful for avoiding misleading statements about IV variable "importance" when there is collinearity among the IVs (and one looks exclusively at the relative size of the standardized regression coefficients: "betas").

There is no way you could calculate Cronbach's alpha before data collection. The statistic is in fact the consistency observed in the data obtained from your study participants.

There is a 10 participants per item method (I am not sure if this is preferred or not) in estimating Cronbach's alpha. So a sample of 64 should be good with the first scale with six items, but not the remaining two scales with 10 and 14 items, respectively.

For regression, sample of 107 could be too small if you have a large number of predictors. Not enough detail on IVs is provided in your question.

I would not recommend calculating the Cronbach's alpha because it is subject to a lot of criticism. Instead, the omega coefficient can be used, which, as previously mentioned, can be calculated in a simple way using JASP, a statistical software that is completely free ( https://jasp-stats.org/ ). Regarding the size of the sample, I recommend you see:

https://stats.stackexchange.com/questions/10079/rules-of-thumb-for-minimum-sample-size-for-multiple-regression

Article A solution to minimum sample size for regressions