Lesson 10: Univariate Analysis

Each question has only one best answer. Circle clearly the letter of the best answer. If you make a mistake, cross out the circle, and write the letter in capitals next to the question. If a question has both a capital letter and is circled, the letter will be considered to be the answer.

 

1.     What is the difference between descriptive and inferential analysis?

a.     Inferential analysis is qualitative and descriptive analysis is qualitative.

b.     Descriptive analysis makes statements about data collected. Inferential analysis makes statements beyond the data collected.

c.      Inferential analysis involves making statements about data you have collected. Descriptive analysis involves describing the world beyond the data you have collected.

d.     Descriptive analysis refers to texts. Inferential analysis is about statistics.

e.     None of the above.

 

2.     Mike is teaching a large popular computer class that enrolls freshmen, sophomores, juniors, and seniors. Mike wants to know how well students’ ages, education levels, and parents’ income predict students’ final grades. What kind of analysis does Mike need to do?

a.     Univariate

b.     Bivariate

c.      Multivariate

d.     Nonvariate

e.     Qualitative

 

3.     Tommy asked 100 respondents if they liked different email systems. For each system he wrote down a 1 if respondents said that they liked it and 0 if they said they didn’t like it. What kind of variable does this create?

a.     Nominal

b.     Ordinal

c.      Ratio

d.     Neutral

e.     Qualitative.

 

4.     Jane conducted a national mail survey of 300 people. One of the pieces of data she recorded was the zip code from which the survey was mailed. What kind of variable is ZIPCODE?

a.     Nominal

b.     Ordinal

c.      Ratio

d.     Neutral

e.     Qualitative

 

5.     What does it mean to collapse a variable?

a.     To eliminate it from further analysis

b.     To take an interval or interval variable and reduce the number of values.

c.      To remove the outliers

d.     To calculate the central tendency

e.     None of the above.

 

6.     What are measures of central tendency good for? 

a.     To measure the degree of heterogeneity of a variable.

b.     To measure the degree of homogeneity of a variable.

c.      To measure the range of a variable.

d.     To measure the minimum and maximum value of a variable.

e.     To measure the typical value of a variable.

 

7.     In a study of younger New Hampshire users, the Keene Daily News  reports that more users browse with Explorer than any other. What kind of statistic is the Keene Daily News reporting?  

a.     Mean

b.     Mode

c.      Median

d.     Standard deviation

e.     Variance.

 

8.     Tom reports that 50% of test scores in his class were above 79 and 50% were below this point. What kind of statistic is Tom reporting?

a.     Mean

b.     Mode

c.      Median

d.     Standard deviation

e.     Variance

 

9.     Lisa records the number of hits per hour for 8 bulletin boards as 1, 3, 5, 26, 54, 57, 59, 75. What is the median?

a.     26

b.     35

c.      40

d.     53

e.     none of the above.

 

10.            The median represents what percentile? 

a.     The first quaratile

b.     The third quartile

c.      The interquartile range

d.     The 50th percentile

e.     It depends on the data.

 

11.            How do you calculate the mean of a set of numbers

a.     Sum the individual scores in a distribution and divide by the number of scores.

b.     Take the smallest number and add the biggest number and divide by two.

c.      Subtract the distribution average from each number and sum the differences.

d.     Subtract the distribution average from each number and square the result. Sum these squared differences and divide by the number of items in the set.

e.     None of the above.

 

12.            In a demographic survey of 20 households, Tom finds 4 households with 4 children, 4 households with 3 children, 4 households with 2 children, 4 households with 1 child, and 4 households with no children. What is the mean number of children per household

a.     1.5

b.     2

c.      2.5

d.     3

e.     5

 

13.            Alex ran a telephone survey during the late morning on a Tuesday. He interviewed people with the following ages: 15, 55, 57, 58, 58, 60, 63, 70. Which age(s) is(are) the outliers?

a.     15

b.     17 and 70

c.      58

d.     58 and 58

e.     None of the above.

 

14.            In a distribution that is skewed to the right, you should expect that:

a.     There are two modal responses

b.     The median equals the mean

c.      The median is smaller than the mean

d.     The median is bigger than the mean

e.     The mode is bigger than the mean.

 

15.            On examining the distribution of one of his interval-level variables, Robert discovers that the distribution around the mean and median are quite symmetrical. What should he use as an estimator of the central tendency?

a.     Mean

b.     Median

c.      Mode

d.     Range

e.     Standard deviation.

 

16.            What is the standard deviation?

a.     It is a formal measure of skewedness.

b.     It results from subtracting the mean from each observation, squaring and summing the differences, and dividing by the sample size minus 1.

c.      It is the average squared deviation from the mean.

d.     It is a measure of how much, on average, the scores in a distribution deviate from the mean.

e.     None of the above.

 

17.            John asks 100 people to report their average income. The mean income for the sample is $37,500, and the variance is $2,500. What is the standard deviation for John’s sample?

a.     $25

b.     $50

c.      $500

d.     $2,500

e.     You cannot tell from the available information.

 

18.            Tim wants to test whether the average age of his sample of 200 respondents is the average age of the population from which his sample was drawn. What is Tim’s null hypotheses?

a.     Any difference between the sample mean and the population mean is produced by chance.

b.     There is a difference, not due to chance, between the sample mean and the population mean.

c.      The sample mean will vary less than the population mean.

d.     The sample mean will vary more than the population mean.

e.     Both a and c above.

 

19.            Jenny set her experiment significance level at 0.01. This means that:

a.     It if the sample is 1% bigger than the mean of the population, she cannot reject her null hypothesis.

b.     If the sample is 1% bigger than the mean from the population, she can reject her null hypothesis.

c.      If a mean or a proportion from a sample is likely to occur more than 1% of the time, then she cannot reject her null hypothesis.

d.     If a mean or a proportion from a sample is likely to occur more than 1% of the time, then she can reject her null hypothesis.

e.     It is just an arbitrary cutoff Jenny picked to report results.

 

20.            Rejecting the null hypothesis by mistake is what kind of error?

 

a.     A Type I error.

b.     A Type II error.

c.      A Type III error.

d.     A sampling error.

e.     A standard error.

 

21.            What should an investigator do to avoid a Type I error?

a.     Set the alpha level of significance at a bigger number (e.g. 0.1)

b.     Set the alpha level of significance at a very small number (e.g. 0.001)

c.      Reduce the sample size.

d.     Increase the sample size.

e.     None of the above.

 

22.            What is a standard error?

a.     The amount of error distributed around the mean sample.

b.     The standard deviation divided by the square root of the sample size.

c.      The amount of error in poor measurements.

d.     The amount of error we make in estimating a population parameter from a sample statistic.

e.     Both b and d are correct.

 

23.            Homero is studying online trading and income. Homero identifies a very small sample (N=10) of traders with low incomes. He wants to know if this sample is significantly different from the average since he knows the US average income. What statistic should he use to test if there are differences?

a.     95% confidence intervals.

b.     Student’s t.

c.      Standard deviations

d.     Univariate chi-square

e.     None of the above.

 

24.            What does a Z-score tell you?

a.     The sum of the standard deviations.

b.     The same thing as a t-statistic, except for larger samples.

c.      The degree to which a population is heterogeneous, using standardized units.

d.     How much standard error there is in any distribution.

e.     How far, in standard deviations, a real score is from the mean of the distribution.

 

25.            What does a univariate chi-square test do?

a.     It tests whether the means of two distributions are significantly different from one another by chance along.

b.     It tests whether the distribution of a series of counts is likely to occur by chance.

c.      It tests whether the mean of a sample is likely to include the mean of the population from which the sample was taken.

d.     It tests interval level associations.

e.     None of the above.

 

26.            T F It is impossible to calculate a median with interval-level data.

a.     True

b.     False

 

27.            T F When data are normally distributed, the mean is the best indicator of central tendency.

a.     True

b.     False

 

28.            T F Whenever possible, investigators should avoid collecting grouped data when they can collect interval-level data.

a.     True

b.     False

 

29.            In samples with a normal distribution, the mean, median, and mode are the same.

a.     True

b.     False

 

30.            T F Most real data is normally distributed.

a.     True

b.     False

 

31.            T F If a population is homogeneous with regard to some variable, then the standard deviation of the mean of that variable should be small.

a.     True

b.     False

 

32.            T F Use a two-tailed test when you are interested only in whether the magnitude or some stat