Project3Guidebook2020.pdf

1

Social Science Statistics STA2122.501 ● ONLINE

Project 3: Comparing Global Values and Attitudes

Contents: • Description and Required Materials (Page 1) • Project Instructions (Page 2) • Grading Rubric (Page 4) • Project Example (Page 6) • APPENDIX A: Variable Codebook (Page 8) • APPENDIX B: SPSS Analysis Instructions (Page 11)

PROJECT 3 DESCRIPTION

For this project, you are being asked to produce a report on the differences in a social value related to a UN Sustainable Development Goal (SDG) using data comparing the US and another country/region (of your choice). Specifically, you’ll be asked to select a country/region of choice and a social value variable, do a little background research, assert a hypothesis, report descriptive statistics, perform a t-test, make a decision with respect to that hypothesis, discuss whether or not the results make intuitive sense, and related back to a UN SDG. This project embodies all of the learning goals and objectives as listed in the syllabus.

1. the vocabulary and symbols used in social statistics 2. how to measure variables and test relationships at different levels 3. the basics of descriptive and inferential statistics 4. to become critical consumers of statistical information 5. about global systems and issues and associated dimensions (e.g., historical, political, economic, social,

cultural, environmental, technological) 6. to analyze global interrelationships and interdependencies across place and time 7. to formally report findings from statistical analyses

REQUIRED MATERIALS

For this project, you will need the following items: • Your NOTES from lessons 1 through 4; 8.1 and 8.2 • A word processing program to type up your final REPORT (e.g., Microsoft Word) • The SPSS data analysis program • The Project 3 data file (Project 3 STA2122 Class Data 2020.sav) posted to Canvas • The VARIABLE CODEBOOK (APPENDIX A) containing a list of the variables and values you may use • The SPSS ANALYSIS INSTRUCTIONS (APPENDIX B) which explain how to use SPSS

University of South Florida Instructor: Dr. Erica L. Toothman

Email: [email protected]

2

PROJECT INSTRUCTIONS

NOTE: The variables you choose to examine in this project do not have to be related to those you presented in Project #1 or #2.

An independent-samples hypothesis test helps us determine if two groups (for example, cats and dogs) substantively differ with respect to a social value as measured by an interval-ratio variable (for example, feelings about lasagna measured on a scale from 1 to 10). For this project, you will be asked to prepare a report that tells us how two groups (for example, the US and Spain) differ with respect to one social value variable related to a UN SDG of your choice. Stated differently, you’ll be comparing local and global data and relating it back to the Sustainable Development Goals set forth by the United Nations.

Project Organization: 1. Variable List (10 points) Read through the Codebook and see what variable interests you. 2. Background Research (15 points) Select and relate your variable to a UN SDG (or pick another variable). 3. Hypothesis statement (10 points) Tell me how the outcome variable (DV) will differ based on country (IV). 4. Descriptive statistics (15 points) Use statistics to describe the sample data for each variable. 5. Annotated test statistics (15 points) Perform a statistical test and explain the what the stats mean. 6. Hypothesis decision (10 points) Do you reject or fail to reject your null hypothesis? 7. Reflection (15 points) Unpack how your findings fit into the bigger picture. 8. Presentation Quality (10 points) Could a stranger pick up your report and understand it?

Stepwise instructions. For information on how to perform these tests in SPSS, please refer to the Analysis instructions (APPENDIX B). Component Requirements:

1. Variable List. This part is easy – in a couple of sentences, tell us about your two variables of interest. Make sure you label and describe them clearly. Here’s what you need to do first:

a. Review the codebook (APPENDIX A) and select two variables you’re interested in researching and analyzing.

b. Go here to read about the UN SDGs: https://sustainabledevelopment.un.org/?menu=1300 c. Select one dichotomous grouping variable (a.k.a., an independent variable). This variable will help

you sort your class-level data from another country’s data. Include the question language and categories in your write-up.

d. Select one outcome social value variable (a.k.a., a dependent variable). This variable should be something you’d be interested in researching related to any UN SDG. The only requirement is that it’s measured on an interval-ratio level and you will need to do some background research on it. Include the question language and categories in your write-up.

2. Background Research. Here, in a half-page, you should provide us with a small report about your

social variable of interest (e.g., feelings about lasagna) and how it operates in both the United States and another country of your choosing (please check the codebook below for countries with available data). Your critical thinking and this background research should inform a comparative research question. The goal here is to inform the reader beyond reporting the numbers -- you should be able to illustrate how this value is different or similar on local and global levels. You should also illustrate how investigating this variable will enhance our understanding of a particular UN SDG. Your research question should inform your hypothesis.

a. YOU MUST UTILIZE TWO VALID EXTERNAL SOURCES OF INFORMATION HERE.

3

3. Hypothesis statement. The goal of an independent samples t-test is to assess how two different groups vary with respect to one outcome variable. I would like you to clearly write one null and one research hypothesis using statistical language. Please indicate if your hypothesis is one- or two- tailed. If you need to, please include a narration/explanation of the hypothesis. Lesson 8.2 will show you how to set up an independent-samples hypothesis test.

4. Descriptive statistics. In a small paragraph, please provide us with descriptive statistics for your

sample. For each group, please include the following: a description of the frequencies, central tendency, and variability (i.e., the mean, the standard deviation, and the standard error of the mean). Keep in mind that you may be reporting on a variable measured on a scale ‘1-4’ or ‘0-10,’ so what does your obtained mean value REALLY indicate? What does this value say about your population? (i.e., who are you talking about)?

a. Use SPSS to generate these statistics (See APPENDIX B it’s super easy). b. Refer to your notes from lessons 1-4 for how to interpret these values.

5. Annotated test statistics. Perform an independent-samples hypothesis test in SPSS. Report and label the following statistics. For each, include a one-sentence interpretation of what that value means. Please assume equal variances.

§ mean difference § standard error of the difference § the degrees of freedom § the obtained t-statistic § the t-critical § the P value

6. Hypothesis decision. Using appropriate statistical language (double check with the book), state your decision regarding your hypothesis in one sentence. In a few other lines, explain to us what this decision means. Assume your α (alpha) is set at a 0.05 level.

7. Reflection. In a full paragraph, provide an informed reflection. Referring back to your background research (including the research on your UN SDG), your descriptive statistics, and your hypothesis test, tell us about what you found! You can use the following questions to guide you: What does a previous study related to this information tell us? Why do you think you obtained this result? What are the implications of your results for your UN SDG? Is your variable measured in the best way possible? What do your findings say about the sample? Is it generalizable to the globe? If not, then whom? Are your findings consistent with other findings? What importance does your test have in advancing knowledge? Most importantly – what further research needs to be done on this issue?

4

PROJECT 3 RUBRIC

Variable list (10 Points) • Full Credit. Two variables (one dichotomous sorting variable and one interval ratio social value

variable) are selected and expertly described. Each is correctly labeled as dependent or independent. • Half Credit. Two variables are selected but are inappropriate for the assignment, incorrectly

labeled, or poorly described. • No Credit.

Background research (15 Points) • Exceeds expectations. Writing demonstrates a good grasp of subject matter and is ½ page in

length. A correctly-constructed comparative research question is included. This part contains appropriate background research (UN SDG + other resources) for the variable selected. It includes statistical language that is specific, clear, and thorough. A thoughtful reflection is provided about national and international differences on a particular social value. Absent of ethnocentrism and stereotypes.

• Meets expectations. Writing demonstrates a basic grasp of subject matter and is ½ page in length. A comparative research question is included. This part contains somewhat appropriate background research (UN SDG or other resources) for the variable selected. It includes some statistics. A more general reflection is provided about national and international differences on a particular social value. Absent of ethnocentrism and stereotypes.

• Approaches/below expectations. Writing demonstrates little-to-no understanding of subject matter or is too short in length. Missing research question. This part contains inappropriate background research for the variable selected. It may not include some statistics. A lacking reflection is provided about national and international differences on a particular social value. Presence of ethnocentrism and stereotypes.

Hypothesis statement (10 Points)

• Full Credit. Null and research hypotheses appropriately stated using symbols and/or words. Hypotheses are described as being one-or-two tailed. Hypotheses are logical.

• Half Credit. Some problems with stating the hypotheses appropriately. One of the hypotheses may be missing. Hypotheses may be less than logical or incorrectly designated as one-or-two tailed.

• No Credit. Significant problems stating the hypotheses appropriately. One or both may be missing.

Descriptive statistics (15 Points) • Exceeds expectations. All required statistics (i.e., mean, standard deviation, and standard

error/mean) are reported for each group. Each value is clearly described, explained, and critically reflected on in a way that demonstrates mastery of descriptive statistics.

• Meets expectations. All required statistics (i.e., mean, standard deviation, and standard error/mean) are reported for each group. Each value is more or less described and explained in a way that demonstrates a basic understanding of descriptive statistics.

• Approaches/below expectations. Some required statistics (i.e., mean, standard deviation, and standard error/mean) are reported for each group. Some values are more or less described and explained in a way that demonstrates a below-average understanding of descriptive statistics.

5

Annotated test statistics (15 points) • Exceeds expectations. All required statistics have been included and are labeled appropriately.

Interpretations are correct and clear. • Meets expectations. All required statistics have been included and are labeled appropriately.

Interpretations are more or less correct and clear. • Approaches/below expectations. Errors in labeling or some missing statistics or

interpretations.

Hypothesis decision (10 Points) • Full Credit. Correct decision/interpretation of results. Uses technical language and refers to null

hypothesis. Includes reflection on what this decision means. • Half Credit. More or less correct decision/interpretation of results. May or may not refer to null

hypothesis. • No Credit. Incorrect decision and lacks understanding of how to interpret hypothesis testing.

Reflection (15 points) • Exceeds expectations. Above-and-beyond effort. Reflection demonstrates a basic

comprehension of previous research on a particular social issue, its relationships to a UN SDG, and very good comprehension of basic descriptive and inferential statistics. Included are thoughtful ideas about the implications of their analyses and how this work may contribute to existing and future research.

• Meets expectations. Typical effort. Reflection demonstrates some comprehension of previous research on a particular social issue or its relationship to a UN SDG, and basic comprehension of descriptive and inferential statistics. Included are some ideas about the implications of their analyses.

• Approaches/below expectations. Below-average effort. Reflection demonstrates little/no comprehension of previous research on a particular social issue and below-average comprehension of descriptive and inferential statistics. Included are some generalized ideas about the implications of their analyses.

Presentation quality (10 points)

• Check-plus. Report is polished, organized, adheres to standard written English, and provides all applicable external resources. There are paragraphs with spaces, appropriate numbering, and subheadings. Contains minimal spelling and grammatical errors. A person who is not in this class could pick up this report, read it, and understand it.

• Check. Report is clear, adheres to standard written English, and provides all applicable external resources. There are paragraphs with spaces, appropriate numbering, and subheadings. There may be some formatting errors that make the project difficult to read. Minimal spelling and grammatical errors.

• Check-Minus. Presentation is unclear. There are formatting errors that make the project difficult to read. Contains more than three spelling and major grammatical errors.

6

EXAMPLE PROJECT

The following example demonstrates how a PROJECT MIGHT look. The data in this sample is derived from another source and some information has been removed and redacted to reduce copying. In lieu of countries, I use gender as my sorting variable.

----------------------------------------------------------------------------------------------------------------------------------

1. Variable List. The variables I chose to explore in this report are gender and mental health. Gender is my dichotomous sorting variable. It’s measured with two categories that include ‘women’ and ‘men.’ My variable of interest – my outcome variable -- is mental health, which is derived from a question asking how many poor mental health days they have experienced in the past 30 days. Mental health is measured on an interval-ratio and responses can range from 0 to 30.

a. NOTE: You will be required to label which variable is dependent and which is independent.

2. Background Research. I elected to look at gender differences in mental health because there is a lot of literature about these issues. Here is where I might assert a comparative research question. When it comes to gender, women and men may have similar rates of mental illness; however, women may report or experience more depression than their men counterparts, who tend to externalize in the form of antisocial behavior. One of the reasons for this may be because women are more comfortable reporting issues related to mental health, more likely to go to a doctor, and have the vocabulary to explain their emotions (research on gender socialization). Other sources point to the increased social expectations on women to work in the paid workplace, at home and on their appearance.

a. NOTE: Your reflection should be about two paragraphs in length, be informed by two sources, discuss how youre research is related to a UN SDG, and should compare and contrast research in the US with research on another country.

b. SOURCE: http://apps.who.int/iris/bitstream/10665/68884/1/a85573.pdf

3. Hypothesis statement. a. My null hypothesis states that there is not a statistically significant difference in the mean days of

poor mental health between women and men. b. My research hypothesis is that there is a statistically significant difference in the mean days of

poor mental health between women and men. This is a two-tailed hypothesis.

4. Descriptive Statistics. Given my background research, and as you may have detected by my hypothesis, I suspect women’s and men’s mental health to be different. Below I’m reporting the descriptive statistics for both groups (gender) on my dependent variable (mental health days). It appears here that, as I kind of expected, women report more poor mental health days in the past month than men by one day. Here is a sentence where I might discuss what the mean of each group indicates. The mean is the average number of days out of 30 that a person experienced poor mental health. In the next sentence, I’ll provide an interpretation of the standard deviation, which is higher than the mean – this is important and I should tell you what that indicates! And in a final sentence, here is something about the standard error.

7

a. Group 1: Men (they are coded "1" in the data) i. Mean days of poor mental health in past month: 3.43 ii. Standard Deviation: 6.314 iii. Standard Error of Mean: .321

b. Group 2: Women (they are coded "2" in the data) i. Mean days of poor mental health in past month: 4.18 ii. Standard Deviation: 6.644 iii. Standard Error: .351

5. Annotated test statistics. a. Mean Difference: -0.752. A mean difference of -0.752 indicates that ________. b. Std. Error of the difference: .475 c. Obtained t-statistic: -1.582. My t-obtained is -1.582, which indicates that _______. d. Degrees of Freedom: 742. The df for this test is 742, which indicates that _______. e. t-critical: +/- 1.960 (I had to use Appendix C in my textbook!) My t-critical is 1.960, which is

_________ compared with my t-obtained of -1.582. f. p-value: < .0001. A p-value of ".0001" indicates that _______.

6. Hypothesis decision. Is this a significant mean difference between your two groups with respect to your outcome variable? Did women have more poor mental health days than men? Report your decision as whether you reject or fail to reject the null hypothesis and why you came to this decision. (1-2 sentences). Does that necessarily mean that your research hypothesis is definitively true? We are evaluating your ability to read, interpret, and report findings from statistical tests here.

7. Reflection. In a full paragraph, provide an informed reflection. Referring back to your background research, your descriptive statistics, and your hypothesis test, tell us about what you found! You can use the following questions to guide you: What does a previous study related to this information tell us? Why do you think you obtained this result? What are the implications of your results for the UN SDG you’re looking into? Is your variable measured in the best way possible? What do your findings say about the sample? Is it generalizable to the globe? If not, then whom? Are your findings consistent with other findings? What importance does your test have in advancing knowledge? Most importantly – what further research needs to be done on this issue?

8

APPENDIX A: VARIABLE CODEBOOK A codebook contains all possible variables in a given data set. Below is a list containing a subset of questions (i.e., variables) appearing in the most recent wave of Pew’s Global Attitudes Survey that I have given to your class. To read more about the GAS, go to: https://www.pewresearch.org/methods/interactives/international-methodology/all- survey/all-country/all-year. Assume random sampling of adults in each country.

Label Country-based dichotomous sorting variables AUSvUS 1 = Australia; 2 = US

BRAZvUS 1 = Brazil; 2 = US CHIvUS 1 = China; 2 = US GERvUS 1 = Germany; 2 = US JAPvUS 1 = Japan; 2 = US

MEXvUS 1 = Mexico; 2 = US NIGvUS 1 = Nigeria; 2 = US PHIvUS 1 = Philippines; 2 = US SPNvUS 1 = Spain; 2 = US SWEvUS 1 = Sweden; 2 = US

Label Social values variables measured on an interval-ratio level v23 (v23) All things considered, how satisfied are you with your life as a whole these days? Using

this card on which 1 means you are “completely dissatisfied” and 10 means you are “completely satisfied” where would you put your satisfaction with your life as a whole?

• Scale 1 to 10 (1 = completely dissatisfied; 10 = completely satisfied) v55 (v55) Some people feel they have completely free choice and control over their lives, while

other peoplefeel that what they do has no real effect on what happens to them. Please use this scale where 1 means "no choice at all" and 10 means "a great deal of choice" to indicate how much freedom of choice and control you feel you have over the way your life turns out

• Scale 1 to 10 (1 = no choice at all; 10 = a great deal of choice) v59 (v59) How satisfied are you with the financial situation of your household? Please use this

card again to help with your answer (code one number): • Scale 1 to 10 (1 = completely dissatisfied; 10 = completely satisfied)

v95 (v95) In political matters, people talk of "the left" and "the right." How would you place your views on this scale, generally speaking?

• Scale 1 to 10 (1 = Left; 10 = Right)

The variables listed below (v96 to v101) are measured/coded as follows: Now I'd like you to tell me your views on various issues. How would you place your views on this scale? 1

means you agree completely with the statement on the left; 10 means you agree completely with the statement on the right; and if your views fall somewhere in between, you can choose any number in

between. v96 0 = Incomes should be made more equal

10 = We need larger income differences as incentives for individual effort • Scale 1 to 10

v97 0 = Private ownership of business and industry should be increased 10 = Government ownership of business and industry should be increased

• Scale 1 to 10 v98 0 = Government should take more responsibility to ensure that everyone is provided for

provide

9

10 = People should take more responsibility to for themselves • Scale 1 to 10

v99 0 = Competition is good. It stimulates people to work hard and develop new ideas 10 = Competition is harmful. It brings out the worst in people

• Scale 1 to 10 v100 0 = In the long run, hard work usually brings a better life

10 = Hard work doesn’t generally bring success—it’s more a matter of luck and connections • Scale 1 to 10

v101 0 = People can only get rich at the expense of others 10 = Wealth can grow so there’s enough for everyone

• Scale 1 to 10 The variables listed below (v192 to v196) are measured/coded as follows:

Now, I would like to read some statements and ask how much you agree or disagree with each of these statements. For these questions, responses range from a "1" meaning that you “completely disagree,” to

"10" meaning that you “completely agree.” v192 ...(v192) Science and technology are making our lives healthier, easier, and more comfortable.

• Scale 1 to 10 (1 = completely disagree; 10 = completely agree) v193 ...(v193) Because of science and technology, there will be more opportunities for the next

generation. • Scale 1 to 10 (1 = completely disagree; 10 = completely agree)

v194 ...(v194) We depend too much on science and not enough on faith. • Scale 1 to 10 (1 = completely disagree; 10 = completely agree)

v195 ...(v195) One of the bad effects of science is that it breaks down people’s ideas of right and wrong.

• Scale 1 to 10 (1 = completely disagree; 10 = completely agree) v196 ...(v196) It is not important for me to know about science in my daily life.

• Scale 1 to 10 (1 = completely disagree; 10 = completely agree) The variables listed below (v198 to v210) are measured/coded as follows:

Please tell me for each of the following actions whether you think it can always be justified, never be justified, or something in between, using this card. (Read out and code one answer for each statement ):

For these questions, responses range from a "1" meaning “never justifiable,” to "10" meaning “always justifiable.”

v198 ...(v198) Claiming government benefits to which you are not entitled • Scale 1 to 10 (1 = never justifiable; 10 = always justifiable)

v199 ...(v199) Avoiding a fare on public transport • Scale 1 to 10 (1 = never justifiable; 10 = always justifiable)

v200 ...(v200) Stealing property • Scale 1 to 10 (1 = never justifiable; 10 = always justifiable)

v201 ...(v201) Cheating on taxes if you have a chance • Scale 1 to 10 (1 = never justifiable; 10 = always justifiable)

v202 ...(v202) Someone accepting a bribe in the course of their duties • Scale 1 to 10 (1 = never justifiable; 10 = always justifiable)

v203 ...(v203) Homosexuality • Scale 1 to 10 (1 = never justifiable; 10 = always justifiable)

v204 ...(v204) Abortion • Scale 1 to 10 (1 = never justifiable; 10 = always justifiable)

v205 ...(v205) Divorce • Scale 1 to 10 (1 = never justifiable; 10 = always justifiable)

v206 ...(v206) Sex before marriage • Scale 1 to 10 (1 = never justifiable; 10 = always justifiable)

10

v207 ...(v207) Suicide • Scale 1 to 10 (1 = never justifiable; 10 = always justifiable)

v208 ...(v208) For a man to beat his wife • Scale 1 to 10 (1 = never justifiable; 10 = always justifiable)

v209 ...(v209) Parents beating children • Scale 1 to 10 (1 = never justifiable; 10 = always justifiable)

v210 ...(v210) Violence against other people • Scale 1 to 10 (1 = never justifiable; 10 = always justifiable)

11

APPENDIX B: SPSS ANALYSIS INSTRUCTIONS

Independent Samples T-Test In the second portion of Project 3, you will conduct a two-tailed independent samples t-test. This is covered In more detail in the second half of the chapter on hypothesis testing. Once your data are open, click “Analyze”, then “Compare Means”, and finally “Independent Samples T Test” to open the menu to conduct the independent samples t-test.

• You will need to select your test variable and your grouping variable. In this case, I will select H2EE8 as my test

variable and BIO_SEX2 as the grouping variable. BIO_SEX2 is the variable for sex. Click the arrow next to “Test Variables” to select H2EE8 as the test variable, and the arrow next to “Grouping Variable” to define BIO_SEX2 as our grouping variable.

• From here, you will need to click “Define Groups…” to tell SPSS which two categories we will use, to see if there is a difference in our test variable.

• Make sure the bubble next to “Use specified values” is selected. Next, you will need to type the number “1” in

the Group 1 box and “2” in the Group 2 box. What this means, as it will mean in your project, is that Group 1 = male and Group 2 = female. So we will see if there is a significant difference in adolescents’ allowances by gender. Ta da! Click “Continue”.

• Notice that in the parenthetical note next to BIO_SEX2 it says (1 2)? This means it will compare the differences between these two groups. If you do not define the groups, SPSS will not run the test. If you used a variable with more than 2 categories, you would need to make sure you were comparing the two groups you intended to compare. Now click OK.

12

• The output window should display something similar to the following:

• Let’s start off with the first box, labeled “Group Statistics.” • Here we can see the mean, standard deviation, and standard error of the mean for Male and Female. So we can

see that the mean allowance for males was 6.77, while for females it was 7.67. Similarly, we can see the standard deviations and standard errors of the mean, too. Just by looking at the Group Statistics table, we can see that it looks like females earn higher weekly allowances and their standard deviations suggest more variability than for males. This alone cannot tell us if there are significant differences in allowance, between men and women.

• Next, look at the Independent Samples Test table. For the purposes of this assignment, you need to look at the data where equal variances are assumed. This is similar to what we have covered in the memos and the second half of the chapter on hypothesis testing.

• Look at all the information under t-test for Equality of Means test. This just means that SPSS is showing us if the mean allowance amount is actually the same.

• Our obtained t-statistic shows up under the column labeled t. It is -2.817. • The degrees of freedom are labeled as df. df=4777.

Our P-value is labeled as Sig. (2-tailed). • The mean difference is -0.901, and the standard error of the difference is 0.320. • Now look back to the Group Statistics table. We can see that the mean difference suggests that males (Group

1) earn about 90 cents less than females (Group 2) in their weekly allowances. If you label males as Group 2, and females as Group 1, the mean difference would be positive, as would your obtained t-statistic.

• You will need to turn to Appendix C in your textbook to figure out what the t-critical value is. I would select for infinity, since my N is much greater than 120.

• Make your decision about what we should do with the obtained information. In your assignment, I have set your alpha value at 0.05 (α=0.05).

Please let me or your course mentor know if something is not clear or if you have questions.