Job Satisfaction Survey Assignment

Overall Commitment Engagement independent Overall Opportunities Co-worker Supervision Pay Work Level Education Age Observation Gender Read the article about creating Likert scales. You will be creating 8 different scales: work, pay,
supervision, co-worker, opportunities, Overall independent, engagement and commitment. Each scale is
the average of the items within the dimension. For example, the work scale will be the average of the
four items, W1 – W4, for each observation; Overall Independent is the average of the twenty
independent variables for each observation (W1 – O4). Cut and paste (you will need to use the paste
special option) until you have a clean data set that looks like the following: 1 M 30 B N 2.5 3.75 2.5 2.25 3 2.8 3.75 2.5 6 2 F 28 A H 2.25 3 3 4 2.25 2.9 2.25 2.25 4 I recommend that you keep a “clean” copy of this revised/scaled data set saved separately. You will need
to sort the data several times for the following analyses. Begin each analysis with a newly copied version
of the scaled data set. Remember, when you sort the data to sort ALL of the data set, not just the
variable of interest.
1. Descriptive Analysis: Describe the data using basic statistics and percentages. For example, 64%
of the observations are Male. The average Overall Satisfaction rating (dependent) is 6.58. Etc.
Be thorough and use charts, graphs, and tables. Discuss your findings. 2. Correlation analysis: Determine the correlation between each of the 6 independent measures
and the three dependent measures (Work & engagement, work & commitment, work & overall,
etc). Also, compute the correlation between age and the three dependent variables. Discuss
your results.
3. Gender analysis: Is there a statistically significant difference between the genders? Use Excel’s ttest function to test for differences in each of the 9 variables. This will be a two-tailed
hypothesis test: Ho: Wf - Wm = 0; Ha: Wf – Wm ? 0. In case you’ve forgotten how to use the Excel
function, there’s an illustration below. Discuss your results. 4. Age analysis: Is there a statistically significant difference between age and the dependent
variables? Sort each of the dependent data columns into the following age groups: ? 29, 30-39,
40-49, 50-59, and 60-69. Use Excel’s Anova-Single Factor to determine if there is a statistically
significant difference across ages. Use the example below (for engagement) to set up your data
for analysis. Discuss your results. ? 29
2.5
2.5
2.5
Etc. 30-39
40-49
50-59
60-69
3.75
4.25
5
5
3.75
4.25
5
4.25
2.25
4.25
5
Etc.
Etc.
Etc. 5. Education analysis: Can we conclude that there is a statistically significant difference between
level of education and the dependent variables? Sort and set up the data as you did for the age
analysis. Initially, conduct an Anova analysis to determine if there is a difference across
education levels. If there is, then test the following null hypotheses (t-test): Es-Eh? 0; Ea-Eh ? 0;
Eb-Eh ? 0; Eb-Ea? 0; Eg-Eb? 0 (see the t-test in problem 3). Discuss your results. Comment on
whether education level appears to be positively or negatively correlated to each dependent
variable.
6. Employment Level analysis: Repeat the analysis above (ANOVA) using level of employment as
the categorical variable. Use comparable pairings for the t-tests. Comment on whether
employment level appears to be positively or negatively correlated to each dependent variable.
7. Interdependence: Create a 3 x 2 matrix by further categorizing the variables. For the dependent
variable, Engagement, create two categories: unengaged and fully engaged. Code the data as
follows: U if the value is < 3.5; E if the value is ? 3.5. Pick an independent variable (other than
overall because it is used below) and code it as follows: L (low) if it is < 3; M (medium)if it is equal to or above 3 but less than 4; H(high) if it is ? 4. Once you have the data coded, organize it
into a contingency table. You can use Excel’s pivot table to do this, but if you are unfamiliar with
how the function works, it might be faster to manually count the number of times that you see
an M and an E together, for example. The table below illustrates the steps and final table you
will need. You are finally ready to do some analysis! Conduct a chi-squared analysis to
determine if the dependent variable, Engagement, is independent from the independent
variable you have chosen. You will need to repeat this analysis with the other dependent
variable, Commitment, but you can use the same independent variable. Code Commitment as
follows: N (for not committed) if the value is < 3; C if the value is ? 3. Discuss your results. Overall
Observation Indpdnt Engagement
1
2.8
3.75
2
2.9
2.25
3
3.55
4.25
4
3.75
5 Overall
Indpdnt
H
H
H
H Engagement
E
E
E
E E
L
M
H U
10
39
16 31
8
0 8. Conclusions: Summarize what you learned from the above analyses about Job Satisfaction, its
dimensions and its relationship to engagement, commitment, and overall perceived satisfaction.