Please notice this! :( My research team is looking forward to getting any recommendations about the Authorization Request for Scale of the authors, for the approval in relation to the demographic profile of our participants. We haven't find any supporting journals so we are looking for possibilities that are there anyone who happens is a caretaker of the test who purchased and had used it for the deaf and mute employees?
We want to have a Norming for our samples just to have the support that the scale we have used is possible for our participants. Is it true that we need to have to double our respondent before we proceed to norming? e.g. if we have n= 30 then for our norming we should have 30*2 n= 60
These are the scale we are going to use:
- Deborah Dwyer& Daniel Ganster "JOB CONTROL Scale"
- Lennart Sjoberg & Fredrik Lind "WILLINGNESS TO WORK Scale"
- Matthias Jerusalem & Ralf Schwarzer "PERCEIVED SELF-EFFICACY Scale"
We would like to solicit your help by knowing your utmost interest and concern in the value of this research as our source for our data gathering for our Quantitative Research. It will be highly appreciated and be greatly help us. We are looking forward to your support in our growth as future professionals. Thank you very much for your favorable action and God Bless! Sincerely, [email protected]
Hello Victoria,
If you wish to establish norms for your target population (and none currently exist for these measures), 60 cases will not be enough. If, on the other hand, you want to confirm that scores from the 3 measures you named have comparable psychometric properties with your sample as they have with other samples, then a good starting guideline would be to use a sample having at least 5-20 cases per item being evaluated.
More is better, though there is always the phenomenon of diminishing returns as your sample size gets larger and larger. If your population is very limited in size, then you may have to be content with simply getting as many cases as you are able.
For more detail, I'd suggest looking at these sources:
Peterson, N. S., Kolen, M. J., & Hoover, H. D. (1989). Scaling, norming, and equating. In R. L. Linn (Ed.), Educational measurement (3rd ed.) (pp. 221-262). New York, NY: Macmillan Publishing Co.
Kolen, M. J. (2006). Scaling and norming. In R. L. Brennan (Ed.), Educational measurement (4th ed.) (pp. 155-186). Westport, CT: Praeger Publishers.
Good luck with your work!
Hi Sir David! I'm grateful for your comment :( We are asking for your professional help that can help our data to improve for our computations.
- (For reliability) Authorization Request for Scale means- is just that we did a request to authorize if we can use the scale for our respondents which are deaf and mute employees. We can't find any supporting journals that the scales were going to use are available or possible for our deaf and mute respondents because we already try to contact the authors of the test, we are trying to reach or connect them as possible, searching their emails for their approval yet they didn't respond for our request.
- (Scoring) Our concerned is that we are thinking that using 'norming' to interpret the scores in the test for our target samples to represent it as a support in that population. Because, the tool/scales are in a general or self-report measure and has no level to label what range of scores like low, average, and high. Is it okay?
- In norming, our professor told us to double the numbers of participants. For example n=50, if we are going to do norming we are going double it (it will be n=100). In overall of 150 respondents. Is it correct? I hope you get my thoughts :(
Hello Victoria,
1. It is always a good idea to estimate and report score reliability for your own sample(s). That helps others to determine how stable (or volatile) such estimates are across samples. As well, your target population appears to be qualitatively different in some important ways from the kinds of samples the instrument authors likely used when developing their scales.
2. For score interpretation, you could organize a group of experts who understand both: (a) the construct(s) measured by a scale; and (b) your target population, to help establish score thresholds that characterize "high", "average", or "low" levels. However, if your goal is to develop normative tables that are stable estimates of what might be true for the population, more cases is always preferable!
Good luck with your work!
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Probability
- Probability Theory
- Reliability