Grouping variables regarding job security, ANOVA, Post-hoc

See the attached files.

1. The grouping variables are the questions regarding job security.

You are testing to see if there is a mean difference of drinking behavior by their perceptions on job security.

2. The methodology and description of variables sections need to be redone. What kind two can you do with this data?

3. Let me see the ANOVA results (Steps 1-5).

Anova: Single Factor

SUMMARY
Groups Count Sum Average Variance
Very True 177 413 2.333 16.485
Somewhat true 111 389 3.505 21.652
Not very true 65 242 3.723 21.172
Not at all true 36 232 6.444 23.111

ANOVA
Source of Variation SS df MS F P-value F crit
Between Groups 537.472 3 179.157 9.262 0.000 2.628
Within Groups 7446.985 385 19.343

Total 7984.458 388
+++++++++++++++++
post-hoc test will indicate which variable(s) are different.
Post hoc analysis
Tukey simultaneous comparison t-values (d.f. = 385)
Very True Somewhat true Not very true Not at all true
2.3 3.5 3.7 6.4
Very True 2.3
Somewhat true 3.5 2.20
Not very true 3.7 2.18 0.32
Not at all true 6.4 5.11 3.49 2.98

critical values for experimentwise error rate:
0.05 2.60
0.01 3.18

p-values for pairwise t-tests
Very True Somewhat true Not very true Not at all true
2.3 3.5 3.7 6.4
Very True 2.3
Somewhat true 3.5 .0284
Not very true 3.7 .0300 .7505
Not at all true 6.4 0.000 .0005 .0031

Let's look at the post-hoc analysis again.

p-values for pairwise t-tests

Very True
Somewhat true
Not very true
Not at all true

2.3
3.5
3.7
6.4

Very True
2.3

Somewhat true
3.5
.0284

Not very true
3.7
.0300
.7505

Not at all true
6.4
0.000
.0005
.0031

Which variable(s) are different from the others (HINT: look at the yellow cells).

You determined that there is a significant difference between the mean values of the alcohol index and the respondent's financial security (very true = strong, to not at all true = weak confidence).

What does this tell you?