Dissertation Help Needed - SPSS Beginner

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

michael1992
Posts: 3
Joined: Wed Mar 11, 2015 3:07 pm

Dissertation Help Needed - SPSS Beginner

Postby michael1992 » Wed Mar 11, 2015 3:42 pm

Hi there,

I am just looking for some help regarding SPSS, which I am using for my dissertation that is looking at motivation in the workplace, in the context of professional sport.

My research objectives are as follows:
i) Identify key motivators in workplace
ii) Explore difference in these motivators based on career stage/age
iii) Analyse differences in motivators depending on past employment (full-time/part-time history)

The survey questions were as follows:
Section A: Age, Part-time/Full-time, No. years employed, No. years employed full-time.
Section B: Rank top 6 motivational factors from a list of 12 (1-6) in terms of importance.
Section C: What keeps you motivated to play sport professionally? (7 options, could select as many as possible) What career plans do you have for after playing pro sport? Would you consider playing unpaid, after you finish laying professionally?

I think I have the data/variables all inputted and can do the basic frequency calculations but if someone could help me out with which tests would be most suitable then that would be great, as I am finding it very difficult to find the relevant help on line for ranked response answers (such as the question in section B)!

Thanks in advance, and should you need anymore information then please let me know.
iopsych
Posts: 25
Joined: Tue Feb 24, 2015 11:58 pm

Re: Dissertation Help Needed - SPSS Beginner

Postby iopsych » Wed Mar 11, 2015 7:20 pm

Hey,

Based on what you wrote, I think t-test & ANOVA can be used.

For ranked data, you can run non-parametric statistics, which is kind of advanced; or you can transform ranked data to scores. For example, you can assign importance points to motivational factors according to the rank they were given by your respondents. Then use ANOVA / t-test.

I hope this helps.
michael1992
Posts: 3
Joined: Wed Mar 11, 2015 3:07 pm

Re: Dissertation Help Needed - SPSS Beginner

Postby michael1992 » Thu Mar 12, 2015 12:57 pm

Hi iopsych,

Thank you very much for your help. I looked in to both tests and it seems that these are exactly what I need.

However, one question... From my limited knowledge of statistics, I believe that the T-test/ANOVA compares the means or various samples/variables, however with the ranking producing Ordinal data (I think), am I ok to compare? As I have read in certain places that the mean of Ordinal data should not be calculated/compared? Is this an exception?

Thanks again,
M
iopsych
Posts: 25
Joined: Tue Feb 24, 2015 11:58 pm

Re: Dissertation Help Needed - SPSS Beginner

Postby iopsych » Thu Mar 12, 2015 9:15 pm

You are right about not getting means of ordinal data. But in your case you can treat them as continuous variables. Specifically:

For section B, create 12 variables (1 for each motivational factor) and for each participant, type in the rank he/she assigned to the factor in terms of importance. The most important would get a score of 1, the least would get a score of 6, and others would get a score of 0. To increase interpretability, the recode these such that most important is 6, and least important is 1.

For section C, create 7 variables, and give 1 to those participants selected and 0 to those they did not select.

Then you can run several parametric statistics using this data.

You can refer to the link I sent you for more info on how to run each analysis.
michael1992
Posts: 3
Joined: Wed Mar 11, 2015 3:07 pm

Re: Dissertation Help Needed - SPSS Beginner

Postby michael1992 » Thu Mar 12, 2015 11:37 pm

Hi iopsych,

Thanks again for all your help and for sending me on the link, it is very much appreciated.

One question that I can't seem to get an answer for anywhere is the distribution for section B, as each variable will have six answers which equal 0, this makes the data non-normally distributed, which isn't usually acceptable for parametric tests? Does this pose a big problem?

Thanks very much for taking the time to get back to me!
iopsych
Posts: 25
Joined: Tue Feb 24, 2015 11:58 pm

Re: Dissertation Help Needed - SPSS Beginner

Postby iopsych » Sun Mar 15, 2015 12:48 pm

If you refer to my previous post, you will see that each variable will not have six 0 answers. You will create one variable per motivational factor and enter the importance ranks as their ratings. So you will have 12 variables, and each participant (rater) will assign a ranking to each motivational factor. So, if a motivational factor is seen as important by most participants, its score will get high (it will be given many 6s). Using this approach, you can compare motivational factors using t-test.

Hope this helps.

Who is online

Users browsing this forum: No registered users and 6 guests

cron