Struggling with T-Test and Z-Test for Workplace Stress Analysis? Let's Discuss!

Hi,

I am doing my final year project on workplace stress. I have collected the data and used chi-square and ANOVA for data analysis. Now, I want to use t-test and z-test for data analysis. How do I use them? I have used a 5-point rating scale in my questionnaire: Agree, Disagree, Neutral, Strongly Agree, Strongly Disagree. Which one should I use to support my hypothesis?

Example 1: For T-test

Out of 100 samples, 10 of them are department heads.

Null hypothesis: All department heads are feeling stress due to job responsibility.
Alternative hypothesis: Not all department heads are feeling stress due to job responsibility.

Collected data:
Experiencing stress due to job responsibility.
Agree-7, Disagree-1, Neutral-0, Strongly Agree-2, Strongly Disagree-0

Example 2: For Z-test

Null hypothesis: 15% of the employees are feeling stress when working in shifts.
Alternative hypothesis: Less than 15% of the employees are feeling stress when working in shifts.

Experiencing stress when working in shifts.
Agree-6, Disagree-26, Neutral-38, Strongly Agree-10, Strongly Disagree-28

How can I solve this?

Please see [The T-Test](http://www.socialresearchmethods.net/kb/stat_t.php) where you can find some information. The discussions at [One Sample T-Test with Likert Scale Questionnaire - Statistics Help @ Talk Stats Forum](http://talkstats.com/showthread.php?t=8139) will highlight the deficiencies in your approach to using the t-test for the Department Heads' problem. [Likert scale@Everything2.com](http://everything2.com/title/Likert+scale) warns against the use of the t-test when a Likert Scale is used. You can find all this information through a [Google Search](http://www.google.co.uk/search?hl=en&source=hp&q=Likert+sc ale+%2B+t-test&meta=&rlz=1R2SKPB_enGB332&amp ;amp;aq=f&oq=). I am confident that you could have found this information on the web before posting the query. Have a nice day.

Simhan
A retired academic in the UK

Thanks for giving me suggestions. So far, what I have done is everything wrong. How do I interpret Likert scale data? What I have done is added every sample, so I got this: Agree-6, Disagree-26, Neutral-38, Strongly Agree-10, Strongly Disagree-28. What should I do now?

User input: Can you tabulate all the results on an Excel spreadsheet and send it to me? Use 5= Strongly agree and 1= Strongly Disagree sort of codes when entering data for each of the respondents for each of the questions. Questions at the top of the columns and the respondents in each of the rows. If the data is confidential, please drop me a private message, and I will give my email ID. Have a nice day. Simhan

What I have done is similar to percentage analysis. I didn't use an Excel sheet; I have done everything manually. For each statement in my questionnaire, I got results similar to this: Agree-6, Disagree-26, Neutral-38, Strongly Agree-10, Strongly Disagree-28.

I used this data for the analysis of hypotheses with ANOVA and chi-square. For example, the relationship exists between the number of family members and employees' experience of stress when their expenses exceed their income. For the statement "I experience stress when my expenses are more than my income," I obtained these results: Agree-38, Disagree-6, Neutral-26, Strongly Agree-28, Strongly Disagree-2.

I categorized the data according to the number of family members as shown below:

Family Members | Strongly Agree | Agree | Neutral | Disagree | Strongly Disagree
<=3 ............. 2 ................... 9 .......... 2 .......... 1 ................ 1
4 .................. 11 ................. 22 ......... 8 ......... 3 ............... 0
>4 ................. 15 ............... 7 .......... 16 ......... 2 ............... 1

Is this correct?

How do I apply these codes to my statements?
Strongly Disagree - 1
Disagree - 2
Not Sure - 3
Agree - 4
Strongly Agree - 5

Yes, using such codings, please tabulate the data. Also, read the material at the weblinks provided for analyzing data collected using Likert Scales.

Thank you.

Thank you, sir, for your reply. I have completed my project at India Cements Ltd. and I am now tasked with preparing the report. I am considering conducting a sign test to evaluate the facilities provided by ICL to the employees. I have gathered data using a scale of yes and no options, but I am facing some challenges in analyzing the data.

For the sign test, it is necessary to exclude the "no" option and only consider the "yes" and "no" responses when calculating the sample size (n value). Could you please explain how to determine the z value for this analysis? I would appreciate a detailed explanation of the procedure for conducting the sign test. Thank you.

Hello... You are a super moderator. I suppose you can help me. I'm doing a project on stress management in an IT firm. Could you tell me on which things I should focus? How should I carry out the research and what should I do as the first step? Thanks in advance. Sona

Hi, someone help me. How do I do a chi-square analysis using the collected data? I used a five-point scaling, but I don't know how to use the chi-square method to interpret my data.

Karthi