Psyc 21621: Quantitative Methods I

Practice Final ExamMike Moore

READ THE PARAGRAPH BELOW AND SIGN TO INDICATE THAT YOU HAVE READ IT

POSTING OF GRADES: I understand that by coding my Kent State University Student Number or any other number, in the space entitled “Student Number” on the answer sheet for this examination, I give permission for my score on this examination to be posted to public view in association with the number which I code. I understand that if I enter nothing in that space, my score will not be posted to public view in connection with anything that may identify me as an individual.

______

SignatureName Printed

…………………………………………………………………………………………………………………

INSTRUCTIONS (BE SURE TO READ):

  1. Fill out the answer form with your name coded. Code your student number if you agree to have your score posted as indicated above; otherwise leave that space blank. If you are not coding your student number, write it legibly in the margin of the answer sheet.
  1. Please write your name on both your scantron and the test form so you can get credit for the problems in Section II (non-multiple choice), and be sure to return both forms to me.
  1. Answers must be filled out in pencil.
  1. DO NOT bubble in your answers for the questions in Section II, write them on the test.
  1. Be sure to both bubble in your answers on the answer sheet as well as circle your answer choice on this exam (except for Section II, see above). This way if there is a scanning error it can be resolved without issue.
  1. There is one best answer for each question.
  1. Talking or gesturing to another student, looking at another student’s paper, or having notes or texts visible will be considered evidence of cheating. Do not engage in these or any other behavior which could be interpreted as cheating. Keep your eyes on your exam paper.

………………………………………………………………………………………………………………....

Section I: Multiple Choice (Each question is worth 1 point)

  1. The analysis of variance differs from a t test for two independent samples because
  2. the t test is limited to 2 samples
  3. the analysis of variance can handle multiple samples
  4. they test different hypothesis
  5. both a) and b)
  1. The analysis of variance assumes that
  2. observations are random
  3. observations are independent
  4. subjects are related
  5. there are equal numbers of observations in each group
  1. If the null hypothesis in the analysis of variance were true
  2. the variances would all be the same
  3. the sample means would all be the same
  4. the population means would all be the same
  5. every subject would have the same score
  1. Mean squares are closest to
  2. means
  3. variances
  4. standard deviations
  5. F’s
  1. Unequal sample sizes in a one-way analysis of variance are
  2. a major problem
  3. a minor inconvenience
  4. something to be avoided at all costs
  5. something that cannot happen
  1. The Bonferroni procedure controls error rates by
  2. operating at a reduced level of α
  3. operating at an increased level of α
  4. not allowing you to make any comparisons unless the F is significant
  5. changing the error term
  1. A factorial analysis of variance has
  2. more than one dependent variable
  3. more than one independent variable
  4. every level of every independent variable paired with every level of every other independent variable
  5. both b) and c)
  1. In a factorial design a cell is
  2. the combination of a row and a column
  3. the smallest number of subjects who were treated alike
  4. another name for a simple effect
  5. both a) and b)
  1. Which of the following is notanadvantage of factorial designs over one-way designs?
  2. they allow for greater generalizability
  3. they allow us to test an interaction
  4. they make it easier to deal with unequal sample sizes
  5. they give us greater economy
  1. The main effect of a variable is
  2. the effect of that variable controlling for another variable
  3. the effect of that variable averaged over the levels of other independent variable(s)
  4. part of the interaction effect
  5. none of the above
  1. The difference between a point biserial coefficient and a normal Pearson correlation coefficient is that
  2. a point biserial correlation is based on two continuous variables
  3. a point biserial correlation is based on one dichotomous variable and one continuous variable
  4. a point biserial correlation is based on two dichotomous variables
  5. the kind of variable have nothing to do with the issue
  1. A negative relationship between speed and accuracy tells us that
  2. the slowest responders are always the most accurate
  3. the fastest responders are always the most accurate
  4. on average, slower responders are more accurate than faster responders
  5. speed is a virtue
  1. A curvilinear relationship is one in which
  2. one variable always increases as the other increases
  3. as X increases, Y will increase and then level off or fall
  4. as X increases, Y will decrease and then level off or fall
  5. both b) and c)
  1. A newspaper headline writer found that the more adjectives she put in the titles of her articles, the greater the number of newspapers that were sold that day. This relationship between number of adjectives and newspaper sales must be
  2. significantly positive
  3. significantly negative
  4. monotonic
  5. positive

Section II: Problems* (Each question is worth 25 points)

*Remember to show all (if any) work, partial credit will be given

*DO NOT bubble your answers to this section on the scantron

HELPFUL FORMULAS:

s2 = ORs = √(s2)

For Fisher’s LSD:

Cohen’s d =

Reminder: Cohen’s conventions for d – small = .3, medium = .5, large = .8

  1. What effect does smoking have on cognitive performance? Spilich, June, and Renner (1992) asked nonsmokers (NS), smokers who had delayed smoking for three hours (DS), and smokers who were actively smoking (AS) to perform a cognitive task that required the subject to read a passage and then to recall it later. The data follow, where the dependent variable is the number of propositions recalled from the passage. Test at p < .01.

Nonsmokers (NS) / Delayed Smokers (DS) / Active Smokers (AS)
27 / 48 / 34
34 / 29 / 65
19 / 34 / 55
20 / 6 / 33
56 / 18 / 42
35 / 63 / 54
23 / 9 / 21
37 / 54 / 44
4 / 28 / 61
30 / 71 / 38
  1. State symbolically and in words what your H1 and Ho are.
  2. What are your df’s (you should identify 3 df’s)? What is your critical F?
  3. What is the value of your observed F?
  4. Based on the above information, would you reject H0,or fail to reject it? Why? What would your conclusion be?
  5. Run a post-hoc test (Fisher’s LSD) to examine the specific differences between groups NS and AS. What conclusion can you draw from this test?
  6. Calculate Cohen’s d (an estimate of effect size) for the F-statistic that you calculated in part (c). Does this information alter your conclusion from part (d)? How?

Final Exam Answer Key

  1. D
  2. B
  3. C
  4. B
  5. B
  6. A
  7. B
  8. A
  9. C
  10. B
  11. B
  12. C
  13. D
  14. D

(a)H1 = At least one of the three groups will be different from one of the other two - μ1 ≠ μ2 ≠ μ3; Ho = All of the three group means are equal - μ1 = μ2 = μ3

(b)dftotal = 29, dfgroup = 2; dferror = 27

Critical F = 5.49

(c) Descriptives

VAR00002

N / Mean / Std. Deviation / Std. Error / 95% Confidence Interval for Mean / Minimum / Maximum
Lower Bound / Upper Bound
Nonsmokers / 10 / 28.5000 / 13.73762 / 4.34422 / 18.6727 / 38.3273 / 4.00 / 56.00
Delayed Smokers / 10 / 36.0000 / 22.34079 / 7.06478 / 20.0184 / 51.9816 / 6.00 / 71.00
Active Smokers / 10 / 44.7000 / 13.88884 / 4.39204 / 34.7645 / 54.6355 / 21.00 / 65.00
Total / 30 / 36.4000 / 17.85111 / 3.25915 / 29.7343 / 43.0657 / 4.00 / 71.00

ANOVA

VAR00002

Sum of Squares / df / Mean Square / F / Sig.
Between Groups / 1314.600 / 2 / 657.300 / 2.239 / .126
Within Groups / 7926.600 / 27 / 293.578
Total / 9241.200 / 29

(d)Fail to reject Ho because the criticalF exceeds the observed F and conclude that smoking has no effect on cognitive performance.

(e)t = -2.11, Critical t (two-tailed) = 2.052, (one-tailed) = 1.703. We can conclude that the AS group has significantly better cognitive performance than the NS group.

(f)d = .58

It allows us to infer that low sample size resulted in our non-significant p-value and that our IV had a significant effect on our DV.