MGMT 6272 Final
Rules:
1. Take the exam the same as if you were taking it in class.
2. Do not take it with someone else, or discuss it with someone else.
3. Do not look up anything while taking the exam or copy material from any source, including the internet.
Name: ______
1. Twenty Likert scales were summed: A person scoring 80 out of 100 possible (the most favorable) indicates:
a) the person has a favorable attitude toward the object
b) the person has an unfavorable attitude toward the object
c) the person has a favorable attitude compared to the average score of 40
d) the person’s attitude is twice as favorable as another person with a score of 40
2. Which data gathering method has the least structure?
a) face-to-face b) mail c) telephone d) e-mail
3. Which of the following is not an advantage of a personal interview?
a) flexibility b) response rate c) use of different modalities
d) more likely to get people to answer sensitive questions
4. Which of the following is not a measure of reliability?
a) test-retest b) Likert’s Technique c) Kuder-Richardson d) Cronbach’s
5. Which of the following could not be tested with a t-test?
a) one group compared to a hypothesis b) comparing the mean of two groups
c) comparing the mean of three groups d) comparing two measures from the same group
6. Which of the following is true:
a) a test may have validity, but not reliability b) a test may have reliability, but not validity
c) both a and b are false d) both a and b are true
7. Which of the following is false?
a) multiple-choice questions tend to bias results by the order in which the alternatives are given
b) an important advantage of open-ended questions is that respondents may answer them in their own words
c) because respondents are able to better clarify their answers with open-ended questions, the results are easily coded
d) both a and b are false
8. A researcher measured the average income of Iowans and found that income was larger than a computer program could handle, so she divided each income by 10. She has just created what type of scale?
a) ratio b) interval c) ordinal d) inter-categorical
9. The scientific method does not mention:
a) empiricism b) determinism c) behaviorism d) math as an appropriate tool
10. Information will:
a) change what you know b) determine the value of a message c) help solve problems d) all of these
Identify each of the following by the type of scales (Steven’s Scales: i.e., nominal, ordinal, etc.)
1. _____ Your weight 9. _____ Your social class
2. _____ A Likert scale (large sample)
3. _____ Distance from Waterloo to Mason City 10. _____ Pant size
4. _____ The amount of CO2 created by a power plant
5. _____ A measure of gender
6. _____ A Fahrenheit scale
7. _____ The sum of ten nominal scale
8. _____ The ranking of UNI football team
True or False
1. T/F It is possible to gather secondary data without having to impose upon a respondent’s time and goodwill.
2. T/F A hypothetical construct can only be measured by ordinal and nominal scales.
3. T/F Non-sampling errors will decrease as the sample size increases.
4. T/F “Plus-one” sampling can create a random sample.
5. T/F A study can have external validity without having internal validity.
6. T/F The probability of a Type II error can be calculated if the sample size in known.
7. T/F Descartes believed that secondary data was more important than observation.
8. T/F Association can only be determined when interval or ratio scales are utilized.
9. T/F Simpson’s Paradox explains the relationship between scales and sample size.
10. T/F The purpose of a sampling frame is to determine sample size.
11. T/F The probability of a Type I error can be determined by the results of a study.
12. T/F An object is never measured, only the object characteristics.
13. T/F The standard deviation of a continuum is approximately 1/3rd of the range.
14. T/F Cronbach’s Alpha measures the consistency of questions (or scales).
15. T/F It is possible to create a face-to-face interview so that there are no social interactions.
16. T/F Numbers are not necessary to create a ratio scale.
17. T/F Sampling error is the square of the variance.
18. T/F A quota sample is a type if probability sample.
19. T/F An F ratio is computed by divided one mean by another.
20. T/F The scientific method does not allow God to be used as a causal agent, but it does not assume that there is no God.
Identify the following persons:
1. Who first suggested that natural phenomena can be explained only as the effects of natural causes only?
2. Who created the t-test?
3. Who proposed that theories of science must be disprovable?
4. Who said that knowledge indicated that you could do something with what you know?
5. Who created an index of association for interval data?
6. Who insisted that knowledge be gained by experience and logic and not simply from authority?
Create a short response for each question
1. If you were conducting an experiment using deductive logic, what would you do first?
2. What does the Krippendorff’s alpha measure that Cronbach’s alpha does not?
3. Under what condition would information not be the same as data?
4. Why isn’t convergent validity sufficient to establish construct validity?
5. How would it be possible to have a perfect relationship, but have a regression with an R2 = 0?
6.-7. Answer the following questions about the Central Limit Theorem.
a) This theorem tells us about what distribution?
b) What is related to the sampling error of a single sample in this theorem?
c) What is the mean, median, and mode of this distribution?
d) Give two situations in which this theorem does NOT apply?
8. What is the largest sample ever needed, at a 95% confidence, to create a precision of 5% of the mean?
9. What is a ballot bias? Give an example.
10. What is a reactive measure?
11. What is the relationship between type of scales and sample size?
12. If a chi-square of association is used and the probability of the outcome is less than 0.05, what does that tell you about the relationship being measured? (Do not just say that it is significant.)
13. When is a convenient sample pragmatically the same as a simple random sample?
14.- 15. Look at the results below, explain what the statistic is, and what it found.
Group StatisticsSection / N / Mean / Std. Deviation / Std. Error Mean
Q3 / 1.00 / 18 / .7115 / .17379 / .04096
2.00 / 22 / .7386 / .15326 / .03268
t(38) = -0.525, p = 0603
16. The amount of information in a message is related to:
17. On which scale does a score of 60 mean only that the respondent was 30 units more favorable on an attitude scale as a respondent with a score of 30?
18. What size sample would be recommended for a study of net incomes if the standard deviation of such incomes was $15,000, and you wish to make a decision with only a 0.05 chance of being wrong to within $5,000?
19. The square of the Pearson’s Product Moment Correlation Coefficient establishes what?
20. A coin is thrown 400 times. He comes up “heads” 220 times. (Show work)
a) See if the coin is fair using a chi-sq test.
b) See if the coin is fair using a Z-test.
21. Satisfaction is an hypothetical construct. Define the term operational definition, and give an operational definition of a measure of satisfaction.
22. A researcher takes a random sample of 100. She finds that the mean of a measure of attitude is 50 with a standard deviation of 5. Answer the following questions:
a) What is the standard deviation of the sampling distribution?
b) Would you reject Ho: the true mean = 51? Why? (show work)
23. The null hypothesis states that the true proportion is 0.70. How large a sample would be needed to determine a 95% confidence of that hypothesis to with 0.10?
24. What is wrong with the following question?
How often have you visited this museum in the past year?
a) 0 to 2 times b) 3 to 5 times c) 6 to 9 time d) 10 or more times
25.-26. A mailed questionnaire was sent out to 2,000 people selected at random from a population of 200,000. Only 1,000 people sent back the questionnaire. 40% of the respondents said they would like to try a new product that was described on the questionnaire.
a) What is the response rate?
b) What is the researcher’s best guess of what percent of the entire population would like the product?
c) What is the lowest percent of the population that may like the product?
d) What is the largest percent of the population that may like the product?
27. Why can’t a t-test be used when there are more than two groups?
28. The correct answer for a client who asks what sample size is required is:
29. If a sample is a scale model of the population, then ______can be used to estimate ______.
30.- 31. Look at the figure below. Make two hypotheses about this data, and suggest a test for each hypothesis.
You are NOT required to complete this section. However, if you do, the number wrong will be subtracted from the number correct and that will be added to your final exam score.
Below are a number of statistics, match a statistic with each case.
a. Chi-sq for fit
b. Chi-sq for association
c. Median
d. Z-tests
e. T-tests
f. ANOVA
g. Correlation
h. Regression
i. Multiple regression
j. Standard error
k. Mode
l. Krippendorff’s alpha
1. ______is the correct statistic if you want to know if the number of people coming into a store in one hour has changed from a month ago at the same time.
2. ______is the correct statistic if you want to measure association between nominal variables.
3. ______is the correct statistic if you wanted to predict how many gallons of fuel a plane would take if you knew the number of passengers and the weight of their luggage.
4. ______is the correct statistic if you wanted to know if people in different social classes were the same height. Warner identified six social classes.
5. ______is the correct statistic to use to show if the proportions of abandoned houses in Detroit were no higher than the proportions of abandoned houses in Newark.
6.______is the correct statistic if you wanted to know if Hispanics and Asians had
the same average net income.
7. ______is the correct statistic if you wanted to promote the MBA program by giving the average salary of MBA graduates.
8. ______The standard deviation of a population on a certain measure was 10. A researcher looked at some old data from previous years and found that 200 samples of 100 customers each had been taken over the last decade supposedly from the same population. The standard deviation of the means from the 200 samples was 4. What statistic would allow the researcher to know if these old samples were from the correct population?
9. ______is the correct statistic if you wanted to know if GPA is related to future income?
10. ______Twenty trained coders rated the degree of enthusiasm of student proposals. What statistic measures the degree of agreement between the 20 coders?