What information did you use to make this determination – i.e., what measures of central tendency, dispersion? c. Do the examination results appear to be normally distributed, or skewed? What information did you use to make this determination ( Hint: I would recommend the use of a histogram, with the normal curve superimposed )

Univariate Data Analysis

Instructions:

Exercise I:

Professor Dunaway, the Department Head of the Criminal Justice Department at State U has just given his notorious second examination in his research methods course. The examination results can be found in the uploaded SPSS file Exam II. Now that you have some experience in univariate analysis, it is time to demonstrate your understanding of these procedures by performing some basic procedures in SPSS, and answering some related discussion questions.

1. What level of measurement is the variable exam II?

a. Is this variable discrete or continuous? Why?

2. After performing the relative analysis ( Hint: Frequencies, measures of central tendency, and dispersion) answer the following questions on the performance of the class.

a. Should Professor Dunaway be pleased with his examination results? Why or why not?

b. What information did you use to make this determination – i.e., what measures of central tendency, dispersion?

c. Do the examination results appear to be normally distributed, or skewed? What information did you use to make this determination ( Hint: I would recommend the use of a histogram, with the normal curve superimposed )

d. Can we determine causality from these results? Why or why not?

Exercise II:

The newly-appointed drug czar, Mike Methodone, has been tasked with reformulating our nation’s drug policy. As the lifelong populist, the czar has decided to solicit public opinion on the legalization of the “so-called” less harmful drugs – namely marijuana. You have recently been hired as a member of the czar’s research team. Aside from some of the more popular public-opinion polls (e.g., Roper, Gallop, Pew), you are well aware of the General Social Survey (GSS), as arguably the best source of time-series data on hot-button topics such as capital punishment, abortion, gun-control, and the legalization of drugs. Luckily, participants in the 2016 GSS were asked their attitudes regarding the legalization of marijuana. The variable, WEED (last and only variable in the 2018 GSS dataset), has the following answer categories (0 = “NOT LEGAL”, and 1 = “LEGAL”. Mr. Methodone, knowing that the country is as polarized as ever, hypothesizes that 50% of the population (i.e.,.50) will favor the legalization of marijuana, and 50% will not favor legalization. Fortunately, as a recent graduate from a Master’s Program in Criminology and Criminal Justice, you know just the way to address this issue. In particular, you are testing the null hypothesis that the mean level of support for the legalization of marijuana will be 50% — or in this case, .50 . Hint: You are only analyzing one variable, so based on Chapter 7 in your SPSS text, you should be familiar with the proper analytical procedure to be employed here.

You are to answer the following questions:

1. What is the level of measurement for the variable WEED?

a. Is this variable discrete or continuous? Why?

2. After performing the relevant analyses ( hint: only one analysis is necessary here), what is your final judgement on the null hypothesis that 50% of the population supports the legalization of marijuana? How did you make this determination (what test did you perform, and what tests of significance were employed?)?

a. Based on the results of the tests that you performed to answer question 2, are you of the belief that more or less than 50% of the population supports the legalization of marijuana? Why?

3. Based on the results of your analysis, what recommendations would you offer to the Drug Czar, regarding his drug policy – specifically as it pertains to marijuana?

There will either be two separate assignment drop boxes created in Canvas – syntax drop box, and Class Project V drop box – OR students will be able to submit BOTH files into the same drop box. With regard to the syntax submission, students should include the syntax files for all procedures used in Exercises 1 and 2. There is no need for students to submit an output file, as I should be able to run your syntax files for both exercises and get the necessary output. Only the class project V assignment drop box will be evaluated.

Format:

While there is no specific length/page requirement to this assignment, students are asked to answer the questions comprehensively. Assignments MUST be typed, with 1” margins throughout the document. All students must use 12-point, Times New Roman Font. Students should single-space within each answer (including the letters – e.g., 1a, 1b), but double space between answers (e.g., 1, 2, 3).

• Discrete vs. Continuous: This variable is continuous. While test scores are often reported in whole numbers, they represent a continuum of knowledge where fractional points are possible (e.g., 85.5), and there is a meaningful absolute zero (the absence of any correct answers).

• a. Professor’s Satisfaction: This depends on the Mean and Median. If the mean is significantly below 70 (the traditional passing threshold), he should be displeased. If the mean is high (e.g., 82) with a low Standard Deviation, it indicates consistent class mastery.
• b. Measures Used: To make this determination, use the Mean (average performance), Median (the middle score), and Standard Deviation (the spread). A high standard deviation suggests a polarized class (some did great, some failed), whereas a low one suggests uniform performance.
• c. Distribution and Skewness: By viewing the Histogram with a Normal Curve, look for the “tail.”
  • If the tail points right, it is Positively Skewed (the exam was very hard; most scores were low).
  • If the tail points left, it is Negatively Skewed (the exam was easy; most scores were high).
  • Check the Skewness statistic in SPSS; a value between -1 and +1 typically suggests a relatively normal distribution.
• d. Causality: No, we cannot determine causality. Univariate analysis only describes what happened (the scores). It does not explain why (e.g., did they study more, or was the test easier?). Determining causality requires an experimental design or multivariate regression.

• Discrete vs. Continuous: It is discrete. You cannot be “halfway” legal; a respondent falls into one distinct category or the other with no intermediate values.

• Final Judgement on Null Hypothesis: To test Mike Methodone’s .50 hypothesis, you perform a One-Sample T-Test with a Test Value of .50.
  • If the Sig. (2-tailed) value is less than .05, you Reject the Null Hypothesis. This means the population is not evenly split.
• a. Support Levels: Look at the Mean in your One-Sample Statistics table. Since “Legal” is coded as 1, a mean higher than .50 (e.g., .61) indicates that more than 50% of the population supports legalization.

* Exercise I: Univariate Analysis for Exam II.
FREQUENCIES VARIABLES=ExamII
  /STATISTICS=STDDEV MINIMUM MAXIMUM MEAN MEDIAN SKEWNESS KURTOSIS
  /HISTOGRAM NORMAL
  /ORDER=ANALYSIS.

* Exercise II: One-Sample T-Test for Marijuana Legalization (WEED).
* Testing against the Drug Czar's hypothesis of .50.
T-TEST
  /TESTVAL=0.50
  /MISSING=ANALYSIS
  /VARIABLES=WEED
  /CRITERIA=CI(.95).

* Descriptive stats to see the actual percentage of support.
FREQUENCIES VARIABLES=WEED
  /STATISTICS=MEAN
  /BARCHART FREQ.