Part I. Levels of Measurement & Descriptive Statistics
Instructions: First, identify the IV & DV for each of the following hypotheses. Next, identify the level of measurement for the IV & DV. Then select the appropriate descriptive statistics for each variable in the hypothesis.
Problems:
1. A researcher suspects that women marry at an earlier age compared to men. Age of marriage was operationalized by asking: How old were you when you first married? ____ (state exact age).
a. IV___________________________; DV_________________________
b. Level of measurement:
IV_____________________________________; DV_______________________________________
Before the hypothesis is tested, the research obtains descriptive statistics for each of the variables. What are the appropriate descriptive statistics for the IV & DV.
c. IV ____________________________________________
d. DV ___________________________________________
2. A researcher suspects that women are more likely to favor gun control compared to men. Gun control was operationalized by asking a survey question that reads: Do you favor or oppose a 90 day waiting period for people who want to purchase a gun? The responses categories are strongly favor, favor, oppose, strongly oppose, and no opinion.
a. IV ______________________________; DV________________________
b. Level of measurement:
IV ______________________________________; DV ______________________________________
Descriptive statistics:
c. IV ________________________________________
d. DV _______________________________________
Part II: True or False
1. Inferential statistics describes a single variable in the sample.
2. We test the null hypothesis because there is always a possibility that, just by chance, we have drawn a biased sample.
3. The significance level we set when we conduct a hypothesis test can be thought of as the risk that we are willing to take in being wrong if we say the research hypothesis is supported.
4. A researcher wanted to know if there was a relationship between education and selling sex for money. When she obtained a frequency of the variable "sell sex" she found that 98% of the participants said no they had not sold sex. This means that the variable "sell sex" does not have enough variation to conduct a significance test of the hypothesis.
5. Descriptive statistics is that branch of statistics concerned with generalizing findings from a sample to the population the sample was drawn from.
6. A variable is a characteristic of people, places, or things that changes across cases in a population, while a hypothesis is a tentative statement about the relationship between two or more variables.
8. A researcher obtains descriptive statistics on the variable income (measured as gross income reported on a W2 form). The mean income for the sample is 50,000 and the median is 45,000. This finding tells the researcher that there may have been some high outliers on the variable income.
9. In a quantitative research study the variables in the hypothesis are operationalized in the
findings section of the paper.
10. Researchers typically place the research question or hypothesis in the last paragraph of the literature review.
11. The level of measurement that a variable is operationalized at determines the type of statistical test that can be conducted on that variable.
12. Every day, for a period of one month, Professor Jennings weighs herself on the same scale. The scale reads 90 lbs. each time she weighs. This suggests that the scale is reliable but not valid.
Part III. Short Answer
Instructions: Read the two excerpts below from Cell Press then respond to the questions in the space below each excerpt.
Excerpt 1.
Why malnutrition is an immune disorder
Summary:
Malnourished children are most likely to die from common infections, not starvation. New experimental evidence indicates that even with a healthy diet, defects in immune system function from birth could contribute to a malnourished state throughout life. Researchers speculate that targeting immune pathways could be a new approach to reduce the poor health and mortality caused by under- and over nutrition.
Question: What information do you need to assess whether or not this is a reasonable claim?
Excerpt 2. The following excerpt is from a Web page titled Association of Mature American Citizens.
Executive Summary
Unlawful immigration and amnesty for current unlawful immigrants can pose large fiscal costs for U.S. taxpayers. Government provides four types of benefits and services that are relevant to this issue:
• Direct benefits. These include Social Security, Medicare, unemployment insurance, and workers' compensation.
• Means-tested welfare benefits. There are over 80 of these programs which, at a cost of nearly $900 billion per year, provide cash, food, housing, medical, and other services to roughly 100 million low-income Americans. Major programs include Medicaid, food stamps, the refundable Earned Income Tax Credit, public housing, Supplemental Security Income, and Temporary Assistance for Needy Families.
• Public education. At a cost of $12,300 per pupil per year, these services are largely free or heavily subsidized for low-income parents.
• Population-based services. Police, fire, highways, parks, and similar services, as the National Academy of Sciences determined in its study of the fiscal costs of immigration, generally have to expand as new immigrants enter a community; someone has to bear the cost of that expansion.
The cost of these governmental services is far larger than many people imagine. For example, in 2010, the average U.S. household received $31,584 in government benefits and services in these four categories.
Question 1: How confident are you that the claim made in this excerpt that undocumented immigration is a huge cost with is accurate? Why or why not?
Question 2: How might some researchers counter the claims made in the above excerpt? That is, can you think of other variables (variables other than government benefits, welfare benefits, and public education) that should be examined? How might the addition of these variables counter the claim in the above excerpt? (Explain your answer).
Part IV:
Read the article, "The Relationship Between Childhood Sexual Abuse and Adult Mental Health Among Undergraduates: Victim Gender Doesn't Matter," in the folder titled, "Final Exam Materials" on Black Board. Then, answer the following questions on the article.
1.A Identify and re-state the author's hypothesis regarding Childhood Sexual Abuse (CSA) and depression?
Hypothesis:
1.B How do the researchers expect gender to work? That is, do they think women or men will differ on CSA on depression?
2. How do the authors operationalize the IV?
3. How do they operationalize the DV?
4. Look at the results reported in the very last paragraph of the results section (right above the section "discussion"). What does this paragraph tell you about CSA and psychological distress and gender, CSA and psychological distress? (Do you best to report Chi-Square results to support your response).
Part V: Multiple Choice
Instructions: Circle the one best response to the following questions/statements.
1. A sample is a ___________________.
a. census
b. population
c. sub-set of a population
d. all of the above
2. A correctly drawn probability sample ___________________.
a. is always biased
b. ensures that everyone in the population has the same chance of being selected for the study
c. is representative of the population the sample was drawn from
d. A & C
e. B & C
3. Biased samples are most often used when _________.
a. it is too costly to obtain a probability sample
b. it is too difficult to obtain a list of everyone in a population (e.g., LGBT teens)
c. a researcher wants to conduct a small pilot study
d. all of the above
e. none of the above
4. Qualitative researchers always strive to draw a ____________ sample.
a. biased
b. random
c. unbiased
d. none of the above
5. Which of the following is a reason that we obtain descriptive statistics?
a. To check for data entry errors.
b. To check for high or low values (outliers)
c. To make sure that there is enough variation in the variable
d. A & C
e. A, B & C
6. Inferential statistics allows us to:
a. view the distribution of a single variable in our sample.
b. generalize a finding from the sample to the population.
c. see if we have high or low outliers.
d. none of the above
7. The significance level associated with a statistical test tell us:
a. if we have proof that the research hypothesis is supported.
b. how often we would be wrong if we reject our null hypothesis and say we have support for the research hypothesis.
c. how often we would be wrong if we reject our research hypothesis.
d. all of the above.
Part VI: Interpretation
Instructions: A researcher hypothesized that people who identify as politically liberal are more likely to say that they trust others compared to people who identify as conservative. Use the SPSS data output below to respond to the questions (on the following page) about this hypothesis.
Political Views
CROSS TAB
|
CAN PEOPLE BE TRUSTED * political views Crosstabulation
|
|
|
political views
|
Total
|
|
Liberal
|
Moderate
|
Conservative
|
|
CAN PEOPLE BE TRUSTED
|
CAN TRUST
|
Count
|
160
|
199
|
156
|
515
|
|
% within political views
|
38.1%
|
29.6%
|
29.7%
|
31.8%
|
|
CANNOT TRUST
|
Count
|
228
|
442
|
350
|
1020
|
|
% within political views
|
54.3%
|
65.8%
|
66.7%
|
63.1%
|
|
DEPENDS
|
Count
|
32
|
31
|
19
|
82
|
|
% within political views
|
7.6%
|
4.6%
|
3.6%
|
5.1%
|
|
Total
|
Count
|
420
|
672
|
525
|
1617
|
|
% within political views
|
100.0%
|
100.0%
|
100.0%
|
100.0%
|
Chi-Square Tests
|
|
|
Value
|
df
|
Asymp. Sig. (2-sided)
|
|
Pearson Chi-Square
|
21.782a
|
4
|
.000
|
|
N of Valid Cases
|
1617
|
|
|
|
|
Questions:
1. What does the cross tab tell you about political orientation and trust? (report data)
2. Is the hypothesis supported? How do you know? (Report results of the Chi-Square test).
3. How often would you be wrong if you said that the hypothesis is supported? (Discuss significance in your response.)
Final Question
1. Why can't we ever say that we have proof that a hypothesis is true? (Think about sampling)