Social Studies 201

Winter 2004

Problem Set 5

Due April 7, 2004

1. Shift work and health. The data in Table 1 come from a survey of Canadian workers who work evening and regular shifts. The percentages in the table refer to the percentage of workers, of each characteristic, who report psycho-social difficulties.

For each of married men and women working evening shift, obtain 90 per cent interval estimates for the proportion who report problems with partner.
How large a sample size would be required to estimate the true percentage of any group having high stress, correct to within plus or minus four percentage points, with 95 per cent confidence?
At the 0.05 level of statistical significance, test whether (i) the proportion of men working evening shift reporting high stress exceeds the proportion of men working regular shift that report high stress, and (ii) the proportion of women reporting high stress differs between those working evening and regular shift.
Using these results, comment on the following statement. “For men, the evening shift was particularly associated with psychosocial difficulties. Married men working an evening shift were more likely than those working regular daytime hours to report relationship problems. … On the other hand, women working the evening shift did not report similar psychosocial problems.” (from Margot Shields, “The health of Canada’s shift workers,” Canadian Social Trends, Summer 2003, No. 69, p. 23).

Table 1. Psycho-social difficulties of men and women workers with different types of work schedules. Per cent of each group reporting difficulty

Characteristic of worker and psycho-social problem	Evening shift		Regular shift – per cent
Characteristic of worker and psycho-social problem	Per cent	Sample size	Regular shift – per cent
Men – high stress	44%	137	33%
Married men – problems with partner	36%	93	16%
Women – high stress	41%	149	43%
Married women – problems with partner	29%	99	21%

Source: Margot Shields, “Shift work and health,” Statistics Canada, Health Reports, Volume 13, Number 4, (catalogue number 82-003-XPE), Tables 1 and 2.

2. Television and internet use. The data in Table 2 come from Saskatchewan respondents aged 15-24 surveyed in Statistics Canada, 2000 General Social Survey, Cycle 14: Access to and Use of Information Communication Technology. Use these data to answer the following for Saskatchewan residents aged 15-24:

A researcher claims hours of television use and of internet use at work and at home each average 11 hours per week. Test the researcher’s claim for (i) watching television and (ii) use of internet at work (0.01 significance level). Briefly explain how and why your conclusions differ.

Table 2. Weekly hours of television and internet use, Saskatchewan respondents aged 15-24 who used each service, General Social Survey, Cycle 14, 2000

Variable	Mean hours per week	Standard deviation of hours per week	Sample size
Watch television	13.46	10.18	180
Use internet at home	9.03	9.61	63
Use internet at work	7.60	10.80	15
Use internet at school	3.98	3.89	45

3. Years of education. The data for this question come from the Saskatchewan respondents in the Survey of Labour and Income Dynamics (SLID) 2000, conducted by Statistics Canada. Data refer to the adult population, aged 15 or more.

a. The sample of 3,862 Saskatchewan respondents in the survey reported their mean years of education as 12.44 years, with a standard deviation of 3.14 years. At the 0.02 level of significance, test whether this provides evidence that the mean years of education for all Saskatchewan adults differs from 12 years.

b. Mean education was 13.73 years and the standard deviation was 2.80 years for a small sample of 20 respondents in the SLID survey. Do these results provide sufficient evidence to conclude that the mean education of all Saskatchewan adults exceeds 12 years? (0.05 level of significance).

c. In a few sentences, compare the conclusions of a. and b., commenting on any possible errors in the conclusions.

4. Computer problem. Use the SSAE98 data set (in folder t:\students\public\201) to obtain the following. Print out the results for the a. parts of each question and write the answers to the b. parts on the printout. Alternatively, use Copy objects to move the tables into a Word for Windows file, write the answers to the b. parts there, and then hand in.

1. a. Obtain the following using Analyze-Compare Means-One-Sample T-test.

i. Test whether the mean weekly hours undergraduates spent at religious and spiritual activities differs from that of Saskatchewan residents as a whole. From Statistics Canada, General Social Survey of Canada: 1998 – Cycle 12, Saskatchewan residents report spending a mean of 1.04 hours weekly at religious services, meetings, prayer, etc.

ii. From the same survey, the mean household income for all Saskatchewan residents was $55,107 in 1998. Test whether the mean income for the households or families of students exceeds the provincial mean.

b. Explain the findings from part a. In doing this, make sure you mention the null and research hypotheses, the significance level, and your conclusion.

2. a. Use Analyze-Compare Means-One-Sample T-test to test whether the mean weekly study hours for University of Regina undergraduates (i) exceeds 15 hours, (ii) differs from 16 hours, and (iii) differs from 17 hours. b. Write a short description of the results, commenting in particular on (ii) and (iii).

3. a. Use Analyze-Descriptive Statistics-Explore with Statistics selected and Plots deselected to obtain a 95% interval estimate for mean weekly study hours for all University of Regina undergraduates. b. Explain whether the interval estimate is consistent with results from question 2.

4. a. Obtain a crosstabulation of variable V4 (gays and lesbians treated as married), cross-classified by sex of respondent, using Analyze—Descriptive Statistics—Crosstabs with chi-square selected in Statistics. You may also wish to request the expected values or column percentages in Cells. (It may also help to understand this relationship by using Analyze—Compare Means—Means to request the means of V4 by sex of respondent). b. Explain your findings from the table and the chi-square test, commenting on whether and how male and female undergraduates differ in their responses to V4.