Statistics for the Social Sciences read and reflect

1

Read and Reflect on a Social Science Based Research Paper

Purpose:

The purpose of this task is to develop your statistical literacy. You will learn how to read, interpret and

justify statistical information that are presented in a research article that is selected for this course

component. You will also have the opportunity (optional) to work on this course component with your

peer (another student in your tutorial or in class) in order to practice your statistical communication and

competency skills. This task is similar to what you will encounter in your future workplace. With the

increase of readily available data in workplace environment, we need to be able to make sense of data

and critically evaluate and justify statistical information that we encounter in forms of, for instance,

reports or articles that we may receive.

Selected Article and Useful Information:

We will read and reflect on “Trends in Social Capital in Canada” by Martin Turcotte. This article can

be retrieved from Statistics Canada’s publications page:

https://www150.statcan.gc.ca/n1/pub/89-652-x/89-652-x2015002-eng.htm

Once you click on the above link, you can click on and save the “Full article in PDF” option.

The main source of this study’s data collection is Canada’s General Social Survey (GSS) program. An

overview about this GSS program can be retrieved from Statistics Canada’s page:

https://www150.statcan.gc.ca/n1/pub/89f0115x/89f0115x2013001-eng.htm

From Statistics Canada’s Surveys and Statistical Programs page, you need to refer to three cycles of

this GSS program (2013, 2008, 2003) for more information:

1. General Social Survey – Social Identity, 2013:

http://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SurvId=134876&InstaId=139605&S

DDS=5024

2. General Social Survey – Social Network, 2008:

http://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SurvId=44601&InstaId=30687&SDD

S=5024

3. General Social Survey – Social Engagement, 2003:

http://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SurvId=1390&InstaId=5509&SDDS=

5024

Each of the above links describes the objective of the survey and the ways in which the instrument for

that survey was designed. Moreover, the target population and the method of sampling collection,

which are relatively the same for all these three cycles, is described.

2

Task:

➢ Individually or with another student preferably from your tutorial, you will reflect on this article’s statistical information based on the seven components to consider, which are listed on

pages 3 and 7. These components are divided into three reflection due dates.

➢ For reflections #1 and #3 answer the specific component-questions (in point form is OK) in one-page regular size paper with font size no smaller than 12-point. For Reflection #2, you may

need up to three pages (regular size page with font size no smaller than 12-point). Bring your

reflections with you to your tutorial and hand them in with your tutorial worksheet at the end of

your class to your TA. Your TA will mark your reflection and will hand it back to at the

following tutorial session (or you may follow up with them in their office hours).

➢ If you decide to work with another student in the course, for each due date you and your peer will only write one reflection and submit the pages to one course TA. Please make sure that

your first and last name(s) and your student number(s) are clearly written on each

reflection page that you submit to a course TA.

Assessment:

Most of questions in each component will use the following assessment criteria.

2 = “clearly and adequately addressed the question”

1 = “addressed some parts of the question”

0 = “missed addressing or inadequately reflected on a question”

Reflection # Components Due Date Weight Possible Points

First 1 and 2 in Tutorials on:

Sept 17th – 21st

3% 10 questions: 20 possible points

*2 points for each question

Second 3, 4, and 5 in Tutorials on:

Oct 15th – 19th

4% 18 questions: 120 possible points

*2 points for most questions

Third 6 and 7 in Tutorial on:

Nov 26th – 30th

3% 2 questions: 10 possible points

*5 points for each component

3

Components to Consider: Reflecting on the Selected Article

Part A. Informal Statistical Inference: Chapters 1 to 4.2 of the Textbook

Component 1. Think about the study’s background (the context of the study):

1. Identify its purpose. 2. Identify its variables of interest (e.g., indicators, measurements).

3. Identify the type of its variables (Quantitative [discrete, or Continuous] or Categorical

[Nominal, or Ordinal]).

Component 2. Think about the study’s type and its method of data collection:

1. Specify whether this study is observational or experimental.

2. Who were the target population in this study?

3. What were the total number of respondents in this study?

4. What were the number of respondents in each year that the data was collected?

5. Name the three main instruments used for collecting the data.

6. Briefly describe the three main instruments used for collecting the data.

7. Describe the method of sampling for collecting the data.

Component 3. Think about the study’s Section 1: Trends in Social Networks:

1. Did the percentage of Canadians who reported having no close friends differ in 2003, 2008, and

2013? What were these percentages?

2. Did the percentages of men and women Canadians who reported having no close friends differ

in 2003, 2008, and 2013? What were these percentages?

3. In 2013, compared to persons aged 35 to 44 years, which age group had the largest percentage

with no close friends? Justify your answer with the reported statistics.

4. This report indicates that seniors are especially at risk of having no close friends. “In particular,

34% of seniors who had no close friends stated that their state of health was excellent or very

good, compared with 54% of those who reported having three or more close friends” (Turcotte,

2015, p.5). Based on this information, create a 2-by-2 contingency table, that means a table with

2 rows and 2 columns: 2 x 2. Label its rows and columns.

a. Identify the response variable.

b. Identify the explanatory variable.

c. In the appropriate cells of your table, indicate the reported statistics as well as the implied

statistics (the ones that are not reported but would complete your constructed contingency

table’s information).

d. Sketch a clustered bar chart (it can be drawn by hand) based on the complete information

(the statistics) in your constructed table.

e. Does it appear that (based either on your constructed table or sketched graph) there is an

association between reported number of close friends and reported state of health for senior

Canadians in 2013? That is, do the stated reported health by senior Canadians in 2013 differ

between those who reported having no close friends and those who reported having 3 or

more close friends? Justify your answer.

4

5. Refer to Chart 1: Distribution of the population by number of close friends, 2003, 2008, 2013.

Most Canadians reported having how many (if at all) number of close friends in these three

measured years?

6. This report indicates that proportion of people with three or more friends are on the rise. “In

2013, 75% of Canadians reported having three or more close friends, compared with 70% a

decade earlier (Table 1)” (Turcotte, 2015, p.5). How does this information align with the

graphical display of Chart 1?

7. Which age group had the highest percentage with reported meeting new people in the past

months with plans to stay in contact?

8. This article reports that “the proportion of people with more than 10 other friends was up from

56% in 2003 to 60% in 2013 (Table 2)” (Turcotte, 2015, p.7). How does this information align

with the graphical display of Chart 3?

9. This report indicates that “the percentage of Canadians ages 25 to 34 who had more than 10

“other friends” increased from 58% in 2003 to 70% in 2013” (Turcotte, 2015, p.7). Based on

this information, create a 2-by-2 contingency table, that means a contingency table with 2 rows

and 2 columns: 2 x 2. Label its rows and columns.

a. Identify the response variable.

b. Identify the explanatory variable.

c. In the appropriate cells of your table, indicate the reported statistics as well as the implied

statistics (the ones that are not reported but would complete your constructed contingency

table’s information).

d. Sketch a clustered bar chart (it can be drawn by hand) based on the complete information

(the statistics) in your constructed table.

e. Does it appear that (based either on your constructed table or sketched graph) there is an

association between age of Canadian participants who had more 10 other friends and the

year of study (2003, and 2013)? Justify your answer.

10. In 2013, what percentage of Canadians reported knowing many or most of their neighbours?

Did this percentage increase or decrease from 2008? How about from 2003?

11. This article describes that the likelihood of knowing neighbours and exchanging various favours

varied by age group and province. According to the author what were the hidden/lurking

variables that could contribute to these findings?

12. Refer to Chart 5 on page 9 of this article. Create a contingency table to illustrate whether or not

type of friends (close friends, other friends or acquaintances) differ by number of close friends

and other friends or acquaintances.

13. Describe the obvious trend that you observe in Chart 6. Use and report the approximate

percentages that are presented in this chart to justify your interpretation of this graph.

14. Refer to Table 6 on page 24 of this article.

a. Did the percentage of Canadians who contacted relatives a few times or more a week

change from 2003 to 2013? Justify your answer with the reported statistics.

b. Which age group had the most decrease in their percentage of contact with relatives from

2003 to 2013. Justify your answer with the reported statistics.

c. Which province had the most decrease in their percentage of contact with relatives from

2003 to 2013. Justify your answer with the reported statistics.

5

15. Refer to Chart 8 on page 12 of this article. Create a contingency table to illustrate the

percentage of people who had at least a few friends who came from an ethnic group that was

visibly different from their own, by age group, and year 2003, 2008, 2013.

a. Make a table with 4 rows and 3 columns.

b. Define the age categories in each row; define the year category in each column.

c. Put the joint percentages in each cell of your constructed table. You may use the

approximate percentages that are displayed in this chart.

d. Verify the joint percentages with the statistics provided in Table 7 on page 25.

16. Refer to the section regarding “Factors associated with social networks” on page 12.

a. The author describes that, “after eliminating the effects of individual characteristics such as

age, level of education, and sex, …Canadians were also more likely to have a few other

friends or acquaintances and less likely to have no close friends” (Turcotte, 2015, p.12).

What were the reported predicted probabilities? Use Table 10 to answer this question.

b. It is indicated that “users of social networking sites were more likely than the non-users to

see their friends in-person a few times or more a week (Turcotte, 2015, p.12). What were

the reported predicted probabilities? Use Table 11 to answer this question.

17. This study reports that watching television dose not hinder social contact. Among people having

no close friends, what were the reported predicted probabilities for those who did not watch

television and for those who watched 10 to 19 hours of television per week? Compare these

descriptive statistics.

18. This study reports that people who live in the same city for 10 or more years have more social

contacts. Give short answers to the following questions:

a. For people who resided in their local communities for long period of time, what were the

reported predicted probabilities of having at least three close friends, and more than 10 other

friends? Use Table 10 to answer this question. Compare these descriptive statistics.

b. For people who resided in their local communities for long period of time, what were the

reported predicted probabilities to see, and contact their friends a few times or more a week?

Use Table 11 to answer this question. Compare these descriptive statistics.

c. It is stated that “Living in a Census Metropolitan Area (CMA) may also impact social

networks” (Turcotte, 2015, p.13). Change the word “impact” to more appropriate word(s) in

order to reflect the notion of an observational study.

d. The author reports that “the probability of having contacts with friends from a visibly different

ethnic group was higher for residents of CMA” (Turcotte, 2015, p. 13). What were the reported

predicted probabilities? Use Table 11 to answer this question.

Component 4. Think about the study’s Section 2: Trends in Civic Engagement and Volunteerism:

1. Compare to 2003, about what percentage of Canadians participated in or were member of a

group, an organization or an association in 2013? Report and comment on both these statistics

(in 2003 & in 2013).

2. Refer to Table 12 on page 29 of this article.

a. In 2013, what percentage of Canadians participated in group activities or meetings at least

once a month?

b. In 2013, what were the percentages of men and women Canadians who participated in

group activities or meetings at least once a month? Comment on both these statistics.

6

c. In 2013, did the percentage of Canadians who participated in group activities or meetings at

least once a month vary by age group? If yes, report this range, and indicate the age groups

with the smallest and the highest percentages.

d. In 2013, did the percentage of Canadians who participated in group activities or meetings at

least once a month vary by province? If yes, report this range, and indicate the provinces

with the smallest and the highest percentages.

e. In 2013, did the percentage of Canadians who participated in group activities or meetings at

least once a month differ between 2003 and 2013 among different provinces? If yes, report

the smallest and the highest difference in percentages.

f. The author indicates that there was an increase in the participation in group activities and

meetings for a particular age group between 2003 and 2013. Specify this age group and

indicate the reported statistics. Briefly mention the author’s explanation for this increase.

Component 5. Think about the study’s Section 3: Trends in Feelings of Trust:

1. Refer to Table 13 on page 29 of this article.

a. Compare to 2003, what percentage of Canadians reported that most people can be trusted in

2013? Report and comment on both these statistics (in 2003 & in 2013).

b. In 2013, what were the percentages of men and women Canadians who reported that most

people can be trusted? Comment on both these statistics.

c. In 2013, did the percentage of Canadians who reported that most people can be trusted vary

by age group? If yes, report this range, and indicate the age groups with the smallest and the

highest percentages.

d. In 2013, did the percentage of Canadians who reported that most people can be trusted vary

by province? If yes, report this range, and indicate the provinces with the smallest and the

highest percentages.

2. Refer to Table 14 on page 30 of this article.

a. Compare to 2003, on a scale of 1 to 5, what was the average level of trust in neighbours in

2013? Report and comment on both these statistics (in 2003 & in 2013).

i. Did average level of trust in neighbours differ between men and women? Report and

comment on both these statistics.

ii. Did average level of trust in neighbours vary by age group? If yes, report this range,

and indicate the age groups with the lowest and highest averages.

iii. Did average level of trust in neighbours vary by provinces? If yes, report this range,

and indicate the provinces with the lowest and highest averages.

b. Compare to 2003, on a scale of 1 to 5, what was the average level of trust in strangers in

2013? Report and comment on both these statistics (in 2003 & in 2013).

i. Did average level of trust in strangers differ between men and women? Report and

comment on both these statistics

ii. Did average level of trust in strangers vary by age group? If yes, report this range,

and indicate the age groups with the lowest and highest averages.

iii. Did average level of trust in strangers vary by provinces? If yes, report this range,

and indicate the provinces with the lowest and highest averages.

c. What other basic statistics do you think would be necessary to include in this table?

7

d. Go to SDA @ CHASS – The General Social Survey:

http://sda.chass.utoronto.ca/sdaweb/html/gss.htm

i. Find the General Social Survey on Social Engagement 2008

ii. Click on Data

iii. From the list of variables on the left-side menu, find “Trust and reciprocity”

iv. Expand this “Trust and reciprocity” component to see all its items/questions

v. Select the item associated with:

Trust in (scale of 1 to 5) ...people in your neighbourhood

vi. The variable/item that reflects the statement above is selected; Click on Row

vii. From right-side menu:

• Select “No weight” in Weight

• In Table Options, check off “Summary statistics”

• Click on the Run the Table

viii. Report the summary statistics: Min, Max, Range, Mean, Median, Mode, SD

(Show this part of your SDA output in your answer: Copy and Paste it)

Part B. Formal Statistical Inference: Chapters 4.3 to 9 of the Textbook

Component 6. Think about the study’s statistical methodology:

Refer to Tables 1 to 9, and 12, 13, 14, and 15 (pages 22 to 30 of this study). Think about the

statistical methodologies (e.g., confidence intervals, and/or significance tests) that were used in

this study to report on trends in social capital in Canada.

1. What statistical methodologies were used? Just name them.

2. How or what approaches were used (or should be used) to check for the validity of this

study’s statistical analyses? You may state the necessary assumptions.

Component 7. Think about the expanding the scope of this study:

Think about the ways in which you can expand the scope of this study. Propose one future

research question, quantitative in nature, and a statistical methodology based on what we

covered in the course that can answer your proposed research question.