## Social Studies 306 Fall 1998 Answers for Assignment 2

This file contains some introductory comments on Assignment 2. Following that are suggested responses to the different parts, along with some comments on the assignments that Social Studies 306 students handed in.

• The tables were generally what was requested. The comments were a bit sparse. One of my aims in this class is to encourage students to connect the numbers with text -- to try to show how numerical responses can be summarized in words. With one or two exceptions, I thought the written responses could have been a little more complete.
• Some of you noted that the figures in the tables differed a little from the 1997 Report. In most cases, this was just a rounding difference. I think the report looks better without percentages to the nearest decimal. Note that the figures in this Survey are not accurate to the nearest decimal anyway. There is too much sampling and nonsampling error in this Survey to make results accurate to the nearest tenth of a percentage point. (I might note that the same is probably true for data obtained from most surveys). So, for purposes of the Report, percentages were rounded to the nearest percentage. If a percentage reported to one decimal ends up as exactly half way between two integers, I generally round off to the even integer. For example, I would ordinarily round 28.5% to 28% or 37.5% to 38%.
• The percentage for "meeting people" in the table at the bottom of the first page must be a misprint. I probably copied this incorrectly, and instead of 31% it should be 37%.
• Be careful how use use the word "significant" or "significantly different" when describing results. These words have an exact statistical meaning, referring to the statistical significance of an hypothesis test. Attempt to use some other words when describing differences between values of variables, e.g. "large difference" or "relatively small difference." These are less exact terms, but when looking at differences between numerical values, you are making your own judgment anyway if you have not carried out a statistical test.
• Where values differ, say how they differ. That is, note whether one value is more or less than the other, and whether it is a lot less or a lot more, or whether the difference is relatively minimal. Where there are several sets of differences, note which differences seem greater and which seem less.
```GET
FILE='O:\STUDENTS\PUBLIC\306DATA\746r.sav'.
EXECUTE .
```

## Question 1. a.

FREQUENCIES VARIABLES=social meet .

## Frequencies

Statistics

N
Valid Missing
Socializing 615 131
Meet New People 614 132

Socializing

Frequency Percent Valid Percent Cumulative Percent
Valid 1 Not Important 155 20.8 25.2 25.2
2 Somewhat Important 368 49.3 59.8 85.0
3 Very Important 92 12.3 15.0 100.0
Total 615 82.4 100.0
Missing 9 No response 5 .7

System Missing 126 16.9

Total 131 17.6

Total 746 100.0

Meet New People

Frequency Percent Valid Percent Cumulative Percent
Valid 1 Not Important 386 51.7 62.9 62.9
2 Somewhat Important 204 27.3 33.2 96.1
3 Very Important 24 3.2 3.9 100.0
Total 614 82.3 100.0
Missing 9 No response 5 .7

System Missing 127 17.0

Total 132 17.7

Total 746 100.0

Similarities and Differences. Large percentages of respondents who said they used alcohol said that socializing was somewhat or very important as a reason for using alcohol. All together, 75% of these respondents said socializing was a somewhat or very important reason for consuming alcohol. In contrast, only 37% of these respondents gave meeting new people as a somewhat or very important reason. Almost two-thirds of these respondents (63%) said that meeting people was not an important reason for using alcohol, whereas only one-quarter said that socializing was not an important reason. The two distributions are thus quite different with respondents generally considering socializing to be much more important as a reason for using alcohol than is meeting people.

The percentages in the table in the Report were obtained by grouping together the percentage of respondents (under valid percent) who reported the reason as somewhat or very important. As noted above, the 31% figure for meeting people appears to be a typo.

## Question 1.b.

FREQUENCIES VARIABLES=rgrade12 rgpa drinks /NTILES= 4 /STATISTICS=STDDEV MEAN MEDIAN .
Statistics

N Mean Median Std. Deviation Percentiles
Valid Missing

25 50 75
UR grade point average 504 242 74.09 72.00 7.87 67.00 72.00 77.00
Number of drinks per week 605 141 8.21 5.00 9.43 2.00 5.00 11.00

Frequency Percent Valid Percent Cumulative Percent
Valid 55 12 1.6 1.7 1.7
62 22 2.9 3.2 4.9
67 46 6.2 6.6 11.5
72 90 12.1 12.9 24.5
77 117 15.7 16.8 41.3
82 151 20.2 21.7 63.0
90 257 34.5 37.0 100.0
Total 695 93.2 100.0
Missing System Missing 51 6.8

Total 51 6.8

Total 746 100.0

Frequency Percent Valid Percent Cumulative Percent
Valid 55 10 1.3 2.0 2.0
62 39 5.2 7.7 9.7
67 89 11.9 17.7 27.4
72 146 19.6 29.0 56.3
77 97 13.0 19.2 75.6
82 80 10.7 15.9 91.5
90 43 5.8 8.5 100.0
Total 504 67.6 100.0
Missing System Missing 242 32.4

Total 242 32.4

Total 746 100.0

Number of drinks per week

Frequency Percent Valid Percent Cumulative Percent
Valid 0 70 9.4 11.6 11.6
1 53 7.1 8.8 20.3
2 56 7.5 9.3 29.6
3 43 5.8 7.1 36.7
4 50 6.7 8.3 45.0
5 37 5.0 6.1 51.1
6 36 4.8 6.0 57.0
7 22 2.9 3.6 60.7
8 30 4.0 5.0 65.6
9 18 2.4 3.0 68.6
10 36 4.8 6.0 74.5
11 9 1.2 1.5 76.0
12 25 3.4 4.1 80.2
13 11 1.5 1.8 82.0
14 13 1.7 2.1 84.1
15 9 1.2 1.5 85.6
16 9 1.2 1.5 87.1
17 6 .8 1.0 88.1
18 9 1.2 1.5 89.6
19 2 .3 .3 89.9
20 9 1.2 1.5 91.4
21 5 .7 .8 92.2
23 3 .4 .5 92.7
24 7 .9 1.2 93.9
25 5 .7 .8 94.7
26 5 .7 .8 95.5
27 1 .1 .2 95.7
29 1 .1 .2 95.9
30 5 .7 .8 96.7
32 1 .1 .2 96.9
33 1 .1 .2 97.0
34 2 .3 .3 97.4
35 1 .1 .2 97.5
36 1 .1 .2 97.7
37 1 .1 .2 97.9
38 1 .1 .2 98.0
40 2 .3 .3 98.3
42 2 .3 .3 98.7
43 1 .1 .2 98.8
45 2 .3 .3 99.2
47 1 .1 .2 99.3
54 1 .1 .2 99.5
60 1 .1 .2 99.7
66 1 .1 .2 99.8
67 1 .1 .2 100.0
Total 605 81.1 100.0
Missing System Missing 141 18.9

Total 141 18.9

Total 746 100.0

Some of you noted that there were a lot of missing case for the U of R GPA. But this was as requested in question 42 of the questionnaire. Those students who were in their first semester were not to report a U of R GPA, and this accounts for most of the missing cases.

For alcohol consumption, the number of drinks per week averaged just over eight. Note that the median was only 5 drinks per week, and while there were several students who reported 20 or more drinks per week (about 10%), just over one-half of those who used alcohol reported 5 or less drinks per week.

In the question, there was no intention of relating grades to alcohol consumption, as some of you tried to do. These were just different frequency distributions.

These values may differ from the actual values because (i) the sample was not representative of all undergraduates, (ii) people do not report their responses honestly, but exaggerate or minimize, and (iii) respondents do not know the correct answer (cannot recall or cannot quickly estimate the values). There may be other possible reasons but these would be the main ones.

## Question 1.c.

DESCRIPTIVES VARIABLES=rgpa drinks /STATISTICS=MEAN STDDEV MIN MAX .

## Descriptives

Descriptive Statistics

N Minimum Maximum Mean Std. Deviation
UR grade point average 504 55 90 74.09 7.87
Number of drinks per week 605 0 67 8.21 9.43
Valid N (listwise) 426

The values for the mean and standard deviation are identical to those obtained from the frequencies procedure. This is just an alternative way to obtain these statistics, one that does not produce the whole table. This might be an advantage for DRINKS, the number of drinks per week, since this is such a large table.

## Question 1.d.

(i) The following are the frequency distributions for the seven variables concerning handling alcohol at the top of the second page of the Report. The valid per cents have been rounded to the nearest integer, but other than that these frequency distributions are the ones shown in the Report.

FREQUENCIES VARIABLES=regret suffer relation actions blackout violent law .

## Frequencies

Statistics

N
Valid Missing
Regret Drinking so Much? 591 155
STUDIES SUFFER FROM ALCOHOL? 588 158
RELATIONSHIPS SUFFER FROM ALCOHOL? 587 159
Regret Actions 589 157
Became Violent 588 158
Trouble with law 588 158

Regret Drinking so Much?

Frequency Percent Valid Percent Cumulative Percent
Valid 1 Not at all 189 25.3 32.0 32.0
2 Sometimes 368 49.3 62.3 94.2
3 Often 34 4.6 5.8 100.0
Total 591 79.2 100.0
Missing 9 No response 5 .7

System Missing 150 20.1

Total 155 20.8

Total 746 100.0

STUDIES SUFFER FROM ALCOHOL?

Frequency Percent Valid Percent Cumulative Percent
Valid 1 Not at all 435 58.3 74.0 74.0
2 Sometimes 146 19.6 24.8 98.8
3 A lot 7 .9 1.2 100.0
Total 588 78.8 100.0
Missing System Missing 158 21.2

Total 158 21.2

Total 746 100.0

RELATIONSHIPS SUFFER FROM ALCOHOL?

Frequency Percent Valid Percent Cumulative Percent
Valid 1 Not at all 473 63.4 80.6 80.6
2 SOMETIMES 108 14.5 18.4 99.0
3 A LOT 6 .8 1.0 100.0
Total 587 78.7 100.0
Missing System Missing 159 21.3

Total 159 21.3

Total 746 100.0

Regret Actions

Frequency Percent Valid Percent Cumulative Percent
Valid 1 Not at all 228 30.6 38.7 38.7
2 Sometimes 332 44.5 56.4 95.1
3 A lot 29 3.9 4.9 100.0
Total 589 79.0 100.0
Missing System Missing 157 21.0

Total 157 21.0

Total 746 100.0

Frequency Percent Valid Percent Cumulative Percent
Valid 1 Not at all 417 55.9 70.8 70.8
2 Sometimes 149 20.0 25.3 96.1
3 A lot 23 3.1 3.9 100.0
Total 589 79.0 100.0
Missing System Missing 157 21.0

Total 157 21.0

Total 746 100.0

Became Violent

Frequency Percent Valid Percent Cumulative Percent
Valid 1 Not at all 496 66.5 84.4 84.4
2 Sometimes 83 11.1 14.1 98.5
3 A lot 9 1.2 1.5 100.0
Total 588 78.8 100.0
Missing System Missing 158 21.2

Total 158 21.2

Total 746 100.0

Trouble with law

Frequency Percent Valid Percent Cumulative Percent
Valid 1 Not at all 534 71.6 90.8 90.8
2 Sometimes 46 6.2 7.8 98.6
3 Often 8 1.1 1.4 100.0
Total 588 78.8 100.0
Missing System Missing 158 21.2

Total 158 21.2

Total 746 100.0

(ii) Frequency distributions for the diagrams on the third page, pressure from personal expectations and pressure from competing with other students. In all cases the missing responses have been excluded from the diagrams, so the percentages in the diagrams correspond to the percentages in the valid percent column.

FREQUENCIES VARIABLES=personal compete .

## Frequencies

Statistics

N
Valid Missing
PERSONAL 738 8
COMPETE 739 7

PERSONAL

Frequency Percent Valid Percent Cumulative Percent
Valid 1 No Pressure 6 .8 .8 .8
2 26 3.5 3.5 4.3
3 142 19.0 19.2 23.6
4 362 48.5 49.1 72.6
5 Extreme Pressure 202 27.1 27.4 100.0
Total 738 98.9 100.0
Missing 6 2 .3

9 No response 4 .5

System Missing 2 .3

Total 8 1.1

Total 746 100.0

COMPETE

Frequency Percent Valid Percent Cumulative Percent
Valid 1 No Pressure 151 20.2 20.4 20.4
2 180 24.1 24.4 44.8
3 238 31.9 32.2 77.0
4 126 16.9 17.1 94.0
5 Extreme Pressure 44 5.9 6.0 100.0
Total 739 99.1 100.0
Missing 9 No response 5 .7

System Missing 2 .3

Total 7 .9

Total 746 100.0

(iii) Frequency distribution for the respondents' self reported ethnic or national identity. Note that the numbers in the report have been rounded a bit, for example, the 76.4% who reported themselves as Canadian has been rounded to "three-quarters" in the text at the bottom of the third page.

FREQUENCIES VARIABLES=identity .

## Frequencies

Statistics

N
Valid Missing
Ethnic Self Identity 734 12

Ethnic Self Identity

Frequency Percent Valid Percent Cumulative Percent
Valid 1 Own Ethnic Origin 68 9.1 9.3 9.3
2 Ethnic-Canadian 58 7.8 7.9 17.2
3 Canadian 561 75.2 76.4 93.6
4 Other 46 6.2 6.3 99.9
6 1 .1 .1 100.0
Total 734 98.4 100.0
Missing 9 No response 11 1.5

System Missing 1 .1

Total 12 1.6

Total 746 100.0

## Question 2. a.

MEANS TABLES=drinks cigday BY year /CELLS MEAN COUNT STDDEV .

## Means

Case Processing Summary

Cases
Included Excluded Total
N Percent N Percent N Percent
Number of drinks per week * Year of Program 603 80.8% 143 19.2% 746 100.0%
No. of cigarettes per day * Year of Program 734 98.4% 12 1.6% 746 100.0%

Report
Year of Program Number of drinks per week No. of cigarettes per day
1 First Mean 8.77 1.91
N 237 298
Std. Deviation 9.12 4.51
2 Second Mean 7.92 1.55
N 143 175
Std. Deviation 9.10 4.23
3 Third Mean 7.97 1.10
N 139 166
Std. Deviation 9.42 3.54
4 Fourth Mean 8.33 1.69
N 66 75
Std. Deviation 12.04 4.63
5 Fifth or more Mean 5.00 3.85
N 18 20
Std. Deviation 4.38 7.74
Total Mean 8.23 1.67
N 603 734
Std. Deviation 9.44 4.39

The number of fifth year students is very small, so they will be excluded from this comment. For the others, there is little difference between the mean number of drinks consumed for each year of university. However, first year is highest at 8.8 drinks per week, fourth year is next at 8.33 drinks per week, and second and third year are virtually identical at just under 8.0 drinks per week.

Rather than saying that there is no significant difference in consumption by year, it would be better to say that the differences in consumption by year are small differences, and there is no consistent pattern of change in consumption by year.

A similar pattern does exist for cigarette consumption, with first year highest, followed by fourth year, and with second and third year lower. In this case though, third year is quite a bit lower.

While these patterns do not indicate much difference, it might be worth further investigation to see how or why the years might either differ or not differ.

## Question 2.b.

MEANS TABLES=sthours exthours dephours hwhours volhours relhours BY job /CELLS MEAN COUNT STDDEV .

## Means

Case Processing Summary

Cases
Included Excluded Total
N Percent N Percent N Percent
Study Hours * Hold a job? 708 94.9% 38 5.1% 746 100.0%
Extracurricular Hours * Hold a job? 708 94.9% 38 5.1% 746 100.0%
HOUSEHOLD WORK HOURS * Hold a job? 704 94.4% 42 5.6% 746 100.0%
DEPENDENT HOURS * Hold a job? 707 94.8% 39 5.2% 746 100.0%
VOLUNTARY WORK HOURS * Hold a job? 708 94.9% 38 5.1% 746 100.0%
RELIGIOUS ACTIVITIES HOURS * Hold a job? 710 95.2% 36 4.8% 746 100.0%

Report
Hold a job? Study Hours Extracurricular Hours HOUSEHOLD WORK HOURS DEPENDENT HOURS VOLUNTARY WORK HOURS RELIGIOUS ACTIVITIES HOURS
1 No Mean 17.29 1.90 8.23 5.66 1.67 1.22
N 317 320 318 316 321 321
Std. Deviation 13.77 6.78 29.24 12.57 6.28 5.88
2 Yes Mean 14.51 1.49 3.15 4.25 1.57 .75
N 391 388 386 391 387 389
Std. Deviation 10.63 3.60 13.19 5.12 3.29 1.81
Total Mean 15.76 1.67 5.44 4.88 1.62 .96
N 708 708 704 707 708 710
Std. Deviation 12.21 5.28 22.08 9.25 4.88 4.18

In all cases, the respondents without a job reported higher mean hours spent at the various activities. The largest difference reported was in housework, 8.2 hours per week for those without a job and 3.2 hours per week for this with a job. Those without a job also reported approximately 2.8 hours more study hours per week than those with a job. The other means did not differ greatly, although those without a job averaged higher hours at all activities than did those with a job.

## Question 3.a.

CROSSTABS /TABLES=msupport BY sex /FORMAT= AVALUE TABLES /STATISTIC=CHISQ /CELLS= COUNT COLUMN .

## Crosstabs

Case Processing Summary

Cases
Valid Missing Total
N Percent N Percent N Percent
MSUPPORT * SEX OF RESPONDENT 696 93.3% 50 6.7% 746 100.0%

MSUPPORT * SEX OF RESPONDENT Crosstabulation

SEX OF RESPONDENT Total
1 MALE 2 FEMALE
MSUPPORT 1 Strongly disagree Count 10 9 19
% within SEX OF RESPONDENT 3.9% 2.1% 2.7%
2 Disagree Count 15 4 19
% within SEX OF RESPONDENT 5.8% .9% 2.7%
3 Neither Count 70 118 188
% within SEX OF RESPONDENT 27.2% 26.9% 27.0%
4 Agree Count 118 199 317
% within SEX OF RESPONDENT 45.9% 45.3% 45.5%
5 Strongly agree Count 43 109 152
% within SEX OF RESPONDENT 16.7% 24.8% 21.8%
6 Count 1
1
% within SEX OF RESPONDENT .4%
.1%
Total Count 257 439 696
% within SEX OF RESPONDENT 100.0% 100.0% 100.0%

Chi-Square Tests

Value df Asymp. Sig. (2-sided)
Pearson Chi-Square 23.013(a) 5 .000
Likelihood Ratio 23.027 5 .000
Linear-by-Linear Association 10.556 1 .001
N of Valid Cases 696

a 2 cells (16.7%) have expected count less than 5. The minimum expected count is .37.

Note that there are quite different numbers of males than females in the survey. In terms of numbers of University of Regina undergraduates, males were underrepresented by about 3 percentage points, so that is a problem with the Survey as a whole. But that should not be a particularly greater problem for this cross-classification than for any other part of the survey. What is needed in order to make the analysis of the table easier are row or column percentages. In general, if a variable has only two or three categories, I put this variable in the column and the other variable in the row. Then I request column percentages. This is how I will describe this table. In this case, sex has only two categories, male and female, so this variables is placed in the columns and view of multiculturalism is placed in the rows.

First note that while there are not major differences in the male and female column percentages, there is a statistically very significant chi-square statistic. This means that there is a statistically significant difference in the distribution of male and female responses.

Next note that the middle categories of neither and agree have very similar column percentages. But there are 3.9 + 5.8 = 9.7% of the males who disagree or strongly disagree. This contrasts with only 2.1 + 0.9 = 3.0% of the females who strongly disagree. So there is a greater likelihood of a male disagreeing than there is of a female disagreeing. At the other end of the spectrum, only 16.7% of males strongly agree with multiculturalism, while 24.8% of females strongly agree.

In summary, while there is not a large difference in the distribution of male and female views on multiculturalism, females are generally more supportive of multiculturalism than are males.

## Question 3.b.

CROSSTABS /TABLES=health BY smoke meals /FORMAT= AVALUE TABLES /STATISTIC=CHISQ /CELLS= COUNT COLUMN .

## Crosstabs

Case Processing Summary

Cases
Valid Missing Total
N Percent N Percent N Percent
State of Health * Do Your Smoke? 742 99.5% 4 .5% 746 100.0%
State of Health * Regular Balanced Meals 744 99.7% 2 .3% 746 100.0%

## State of Health * Do Your Smoke?

Crosstab

1 Never 2 Used to 3 Smoke now
State of Health 1 Poor Count 3
4 7
% within Do Your Smoke? .6%
2.9% .9%
2 Fair Count 48 11 24 83
% within Do Your Smoke? 9.5% 11.1% 17.1% 11.2%
3 Good Count 164 44 71 279
% within Do Your Smoke? 32.6% 44.4% 50.7% 37.6%
4 Very Good Count 217 34 25 276
% within Do Your Smoke? 43.1% 34.3% 17.9% 37.2%
5 Excellent Count 71 10 16 97
% within Do Your Smoke? 14.1% 10.1% 11.4% 13.1%
Total Count 503 99 140 742
% within Do Your Smoke? 100.0% 100.0% 100.0% 100.0%

Chi-Square Tests

Value df Asymp. Sig. (2-sided)
Pearson Chi-Square 44.091(a) 8 .000
Likelihood Ratio 45.295 8 .000
Linear-by-Linear Association 25.774 1 .000
N of Valid Cases 742

a 3 cells (20.0%) have expected count less than 5. The minimum expected count is .93.

## State of Health * Regular Balanced Meals

Crosstab

Regular Balanced Meals Total
1 Never 2 Sometimes 3 Often 4 Always
State of Health 1 Poor Count
4 3
7
% within Regular Balanced Meals
1.6% .8%
.9%
2 Fair Count 6 52 21 4 83
% within Regular Balanced Meals 35.3% 20.7% 5.9% 3.4% 11.2%
3 Good Count 6 117 132 25 280
% within Regular Balanced Meals 35.3% 46.6% 36.9% 21.2% 37.6%
4 Very Good Count 2 56 157 61 276
% within Regular Balanced Meals 11.8% 22.3% 43.9% 51.7% 37.1%
5 Excellent Count 3 22 45 28 98
% within Regular Balanced Meals 17.6% 8.8% 12.6% 23.7% 13.2%
Total Count 17 251 358 118 744
% within Regular Balanced Meals 100.0% 100.0% 100.0% 100.0% 100.0%

Chi-Square Tests

Value df Asymp. Sig. (2-sided)
Pearson Chi-Square 104.154(a) 12 .000
Likelihood Ratio 104.426 12 .000
Linear-by-Linear Association 73.957 1 .000
N of Valid Cases 744

a 6 cells (30.0%) have expected count less than 5. The minimum expected count is .16.

For these tables, I placed smoking and well balanced meals in the columns, because these two variables each had fewer categories than did state of health (in the rows). Then I requested column percentages. Note that the chi-square statistic is highly statistically significant in each case, supporting the view that there is a relationship between the two variables in each of the tables. But with so many cells in each table, these tables are more difficult to analyze.

Smoking and Health. In general, there is a tendency for those tho never smoked to report better health. In order to see this, note that 43.1 + 14.1 = 57.2% of those who never smoked report very good or excellent health. For those who used to smoke, but no longer do, 10.1 + 34.3 = 44.4% report very good or excellent health. Finally, of those who smoke now, only 17.9 + 11.4 = 29.3% report very good or excellent health.

Around the middle of the table the differences in column percentages are smaller. But of those who smoke now, 20.0% report poor or fair health, 11.1% of those who used to smoke report poor or fair health, and of those who never smoked, 10.1% report poor or fair health.

In summary, those who smoke now generally report better health than those who used to smoke, and the latter generally consider themselves in better health than those who smoke now. This is a fairly clear relationship between the two variables, with more smoking associated with reports of poorer health.

Well Balanced Meals and Health. This table is even more difficult to analyze but the same approach shows that those who say they eat well balanced meals generally report better health. Note the large percentages in the upper left of the table, with large percentages of those who never or only sometimes each well balanced meals reporting only poor or fair health. In contrast, the upper right of the table has few cases, so that those who often or always eat well balanced meals are unlikely to report poor or fair health.

The respondents who report often or always eating well balanced meals are concentrated in the middle and bottom of the table, reporting good, very good or excellent health. While many of those who never or sometimes eat well balanced meals report themselves in good health, relatively few on these respondents say they are in very good or excellent health.

For this table, do not eliminate categories that have few cases in them, but merge them with adjacent categories. In order to do this on the computer, the RECODE procedure in SPSS will help.

## Question 4.a.

Number of drinks per week by faculty as reported on the first page of the 1997 Report. Note that I used the MEANS procedure. The frequency distribution for drinks in question 1 was very long, and if CROSSTABS were to be used here, this would produce a very large table. Again, the RECODE procedure can be used to produce a table that will not have so many cells.

MEANS TABLES=drinks BY faculty /CELLS MEAN COUNT STDDEV .

## Means

Case Processing Summary

Cases
Included Excluded Total
N Percent N Percent N Percent
Number of drinks per week * FACULTY 604 81.0% 142 19.0% 746 100.0%

 1 ADMIN 9.18 78 12.22 7.75 146 8.16 6.95 57 6.58 10.09 47 10.81 7 7 3.92 11.22 32 16 3.5 2 0.71 8.08 38 8.63 7.48 147 7.62 7.42 43 7.83 3 2 2.83 14.2 5 17.21 8.17 604 9.38

## Question 4.b.

Relationship between reported family income and grade point average. Again the MEANS procedure is used so the table does not have so many cells. Note that while there may be a small positive relationship (correlation of +0.21) it is not statistically significant and is not apparent in the means procedure. The hypothesis that higher family income is associated with higher grade point average is not supported by these data.

MEANS TABLES=rgrade12 rgpa BY income /CELLS MEAN COUNT STDDEV .

## Means

Case Processing Summary

Cases
Included Excluded Total
N Percent N Percent N Percent
Grade 12 grade * Parents Income 695 93.2% 51 6.8% 746 100.0%
UR grade point average * Parents Income 504 67.6% 242 32.4% 746 100.0%

Report
0 Uncertain/Other Mean 79.50 77.00
N 4 4
Std. Deviation 5.00 5.77
1 Less than \$20,000 Mean 79.09 73.13
N 34 23
Std. Deviation 9.85 8.31
2 \$20,000-39,999 Mean 78.61 73.19
N 122 91
Std. Deviation 10.07 8.46
3 \$40,000-59,999 Mean 81.52 74.48
N 143 108
Std. Deviation 7.90 7.60
4 60,000-79,999 Mean 81.80 74.31
N 143 108
Std. Deviation 8.23 8.11
5 \$80,000-99,999 Mean 80.79 73.72
N 80 57
Std. Deviation 9.02 8.31
6 \$100,000-149,999 Mean 81.83 74.24
N 54 42
Std. Deviation 7.99 6.55
7 \$150,000-199,999 Mean 81.47 76.00
N 19 14
Std. Deviation 10.07 9.26
8 \$200,000 plus Mean 76.00 74.82
N 13 11
Std. Deviation 10.91 9.67
9 No response Mean 81.16 74.26
N 83 46
Std. Deviation 9.09 6.64
Total Mean 80.73 74.09
N 695 504
Std. Deviation 8.91 7.87

CORRELATIONS /VARIABLES=inc rgrade12 rgpa /PRINT=TWOTAIL NOSIG /MISSING=PAIRWISE .

## Correlations

Correlations

Pearson Correlation Income in dollars 1.000 .021 .048
UR grade point average .048 .358(**) 1.000
Sig. (2-tailed) Income in dollars . .598 .306