Dear Researchers,
I need your suggestions and recommendations regarding one important question. I am trying to find if there is a significant difference between the two scales.
I have collected data on the acquired skills of our alumni at university and their most required skills at their workplace. They rated these skills on a 7 scale Likert Scale.
I am using SPSS to do my analysis.
Your suggestions will be very welcome since I have to finish my analysis by the end of this month.
Lastly, can I run any regression with my data? I am very new to regression analysis and would love your suggestions.
Thanks a lot!
Regards,
Dzhamshed.
@Dzamshed Your query seems a bit incomplete. Are you looking to compare the responses of two groups on the same questionnaire? If yes then you need to apply independent samples t-test (if assumptions are met for a metric test). Regression is used when you look to measure the relationship(s) between independent and dependent variable(s). For any further assistance, you need to state your case in a bit detail along with your research questions/objectives/hypotheses.
Dear Dzhamshed Salomov,
There are two things in your question.
(1) Using different scales in a questionnaire
(2) Running a regression for the scales.
If you have used different scales for each variable concerned, it is no harm to have them separately and you can run the regression accordingly.
But, for a single variable, you should not have two different scales to weigh the variable. If you have two different scales for a single variable, those scales should be brought up to a single format of measures and then, you can run the regression.
Certain things you have to understand.
a) Univariate analysis is to explore the individual and unique characteristics of a variable.
b) Bivariate and multivariate analyses are to examine the relativity of variables and their impacts on one another.
Correlation analysis is to explore the trending relationship (positive or negative and the degree of the relationship - Weak, moderate or strong) of two variables and regression analysis is to explore the impact of an independent variable on a dependent variable.
Hope the above explanation would give some insights into your question.
Good luck with your studies.
Regards,
S. Senthilnathan.
Imran Anwar Dear Imran,
Thank you for your response. Actually, the respondents are the same group. These are the same number of students who rate their acquired skills at Uni and important skills at work. I want to compare if their acquired ratings are significantly different from the required skills that they rated. Imran Anwar
@Dzamshed In your case, you can apply independent samples t-test (if assumptions are met) else use Man-Whitney U test.
Bilal Mustafa Khan Imran Anwar Thanks for your responses.
I believe that the test which fits my analysis the most is paired t-test since my respondents are the same. However, I am not sure if my data is normally distributed or not. Should I check the normality for each item that were ranked? Or in other words, is there another test that does not assume that the data is normally distributed?
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression