Study Guide



PSY 211 – Exam #4

Mike Hoerger

You may keep this paper copy of the exam when you have finished. You may use a simple calculator during the exam.

The exam is 120 points (3 points per question).

|1 |A “carryover effect” is potential criticism of the within-subject design. |T F |

|2 |The shape of the t-distribution becomes flatter as sample size increases. |T F |

|3 |An alpha level of .10 is typically used in statistical analyses in psychology. |T F |

|4 |Stephen Colbert using Airborne (a useless health pill) was an example of a Type II error. |T F |

|5 |It is possible for Cohen’s d to be “large” and have a result that is not statistically significant. |T F |

|6 |When conducting a between-group t-test, the p-value describes the probability that the observed mean difference is |T F |

| |merely due to chance. | |

|7 |To determine the critical value for a z-test, one must know the degrees of freedom (df) for the sample. |T F |

|8 |In twin studies, the between-group t-test is typically used to compare pairs of twins. |T F |

|9 |When writing an alternative hypothesis, a researcher is describing differences that are expected to occur at the |T F |

| |population level, rather than merely within the sample. | |

|10 |If a correlation has a p-value of 0.40, the correlation is said to be “statistically significant”. |T F |

|11 |In a study comparing a medication to a placebo, the alternative hypothesis would be that there are no differences in |T F |

| |results between the two groups. | |

|12 |When a treatment is more beneficial than expected, it is said to be “iatrogenic”. |T F |

|13 |A Type I error occurs when the researcher incorrectly accepts the alternative hypothesis. |T F |

|14 |The critical value for a one-tailed hypothesis test is less extreme than for a two-tailed hypothesis. |T F |

|15 |For a between-group t-test, Cohen’s d indicates how many standard deviations apart the two sample means are. |T F |

|16 |When conducting a z-test, the critical z-value will vary depending on the sample size used in the study. |T F |

|17 |“Sampling error” means that results can be expected to differ slightly across different groups or samples. |T F |

|18 |When conducting a t-test, choosing a stricter alpha will lead to a critical t-value that is more extreme. |T F |

|19 |The between-group t-test has greater power than the within-subject t-test. |T F |

|20 |“HA” and “H0” are both symbols for the alternative hypothesis. |T F |

|21 |How many of the following analyses allow someone to compare a sample of treated patients to a larger untreated population? |

| |between-group t-test |

| |z-test |

| |single-sample t-test |

| |repeated-measures t-test |

|a) |1 |

|b) |2 |

|c) |3 |

|22 |When writing up results in APA style, it is optional to report a negative sign for this statistic. |

|a) |r |

|b) |z |

|c) |d |

|23 |Cohen’s d is similar to r in the same way that _______ are similar to _______. |

|a) |t-tests; z-tests |

|b) |experiments; surveys |

|c) |Type II errors; Type I errors |

|24 |If Cohen’s d = .47, the effect size is… |

|a) |small |

|b) |medium |

|c) |large |

|25 |A researcher wants to examine whether a brief commercial will influence participants attitudes toward global warming. The |

| |researcher has participants rate their “belief in global warming” on a 9-point scale before and after the commercial. What type|

| |of statistical test would be used to compare participants’ scores before and after the commercial? |

|a) |single-sample t-test |

|b) |within-subject t-test |

|c) |between-group t-test |

|26 |This statistic is an example of an effect size… |

|a) |p |

|b) |r |

|c) |t |

|27 |If r2 = .13, the effect size is… |

|a) |small |

|b) |medium |

|c) |large |

|28 |This type of statistical test could be used to examine gender differences in agreeableness. |

|a) |single-sample t-test |

|b) |within-subject t-test |

|c) |between-group t-test |

|29 |A researcher finds that violent television increases aggression, d = 0.12, t = 6.48, p = .001. What can we conclude about this |

| |particular study? |

|a) |Due to a large sample, a small effect was statistically significant. |

|b) |The result is not trustworthy; it may be due to chance. |

|c) |Several outliers (atypical scores) were likely present in the data file. |

|30 |Below is a correlation table. Note that the asterisks (*) marking significant correlations have been removed. How many |

| |significant correlations does Tanning have with the other variables? |

| |Correlations |

| | |

| | |

| | |

| |89. Leader |

| |ship |

| |44. Tanning |

| |68. Physical Pain |

| |84. Body Satisfac |

| |tion |

| |91. ADHD Symp |

| |toms |

| | |

| |89. Leadership |

| |Pearson Correlation |

| |1.000 |

| |.075 |

| |.045 |

| |.167 |

| |.010 |

| | |

| | |

| |Sig. (2-tailed) |

| | |

| |.019 |

| |.161 |

| |.000 |

| |.751 |

| | |

| | |

| |N |

| |975.000 |

| |975 |

| |975 |

| |975 |

| |975 |

| | |

| |44. Tanning |

| |Pearson Correlation |

| |.075 |

| |1.000 |

| |.040 |

| |.069 |

| |.050 |

| | |

| | |

| |Sig. (2-tailed) |

| |.019 |

| | |

| |.211 |

| |.032 |

| |.122 |

| | |

| | |

| |N |

| |975 |

| |975.000 |

| |975 |

| |975 |

| |975 |

| | |

| |68. Physical Pain |

| |Pearson Correlation |

| |.045 |

| |.040 |

| |1.000 |

| |-.059 |

| |.148 |

| | |

| | |

| |Sig. (2-tailed) |

| |.161 |

| |.211 |

| | |

| |.066 |

| |.000 |

| | |

| | |

| |N |

| |975 |

| |975 |

| |975.000 |

| |975 |

| |975 |

| | |

| |84. Body Satisfaction |

| |Pearson Correlation |

| |.167 |

| |.069 |

| |-.059 |

| |1.000 |

| |-.143 |

| | |

| | |

| |Sig. (2-tailed) |

| |.000 |

| |.032 |

| |.066 |

| | |

| |.000 |

| | |

| | |

| |N |

| |975 |

| |975 |

| |975 |

| |975.000 |

| |975 |

| | |

| |91. ADHD Symptoms |

| |Pearson Correlation |

| |.010 |

| |.051 |

| |.148 |

| |-.143 |

| |1.000 |

| | |

| | |

| |Sig. (2-tailed) |

| |.751 |

| |.122 |

| |.000 |

| |.000 |

| | |

| | |

| | |

| |N |

| |975 |

| |975 |

| |975 |

| |975 |

| |975.000 |

| | |

|a) |0 |

|b) |1 |

|c) |2 |

|31 |William Kingdon Clifford once said, “The danger to society is not merely that it should believe wrong things, though that is |

| |great enough; but that it should become credulous, and lose the habit of __________ and inquiring into them; for then it must |

| |sink back into savagery.” |

|a) |analyzing results |

|b) |testing things |

|c) |studying statistics |

|32 |If Cohen’s d = .68, the effect size is… |

|a) |small |

|b) |medium |

|c) |large |

|33 |A researcher asks smokers (n = 10) and non-smokers (n = 10) to rate their health on a 9-point scale (higher numbers indicate |

| |better health). Their data are as follows. |

| |Smokers: M = 5.23, SD = 1.55 |

| |Non-smokers: M = 6.84, SD = 1.28 |

| |What is the approximate value of Cohen’s d? |

|a) |0.3 |

|b) |1.1 |

|c) |1.6 |

|34 |What is the approximate observed t-value for question #33? |

|a) |1.5 |

|b) |2.0 |

|c) |2.5 |

|35 |On a scale from 0 to 10, the average American reports a happiness level of 7.0. A sample of n = 64 students in PSY 211 report a|

| |happiness level of 6.2 (SD = 2.0). What is the observed t-value for this sample? |

|a) |3.0 |

|b) |3.2 |

|c) |3.4 |

|36 |Anxiety clients (n = 16) report their level of anxiety symptoms before and after a month of therapy, using the Beck Anxiety |

| |Inventory. Their level of anxiety falls by a few points (M = 5.0, SD = 4.0). What is the approximate observed t-value? |

|a) |1 |

|b) |4 |

|c) |5 |

Use the following between-group t-test Output to answer questions 37 through 40 (next page). Note that some sections of the Output are hidden.

|Group Statistics |

| |

| | |Levene's Test for |t-test for Equality of Means |

| | |Equality of Variances| |

| | |

|a) |1 |

|b) |2 |

|c) |3 |

|38 |Although it is hidden, what is the approximate t-value for the t-test involving regretfulness? |

|a) |1.3 |

|b) |1.5 |

|c) |1.7 |

|39 |What is the value for the degrees of freedom (df) for these t-tests? |

|a) |973 |

|b) |974 |

|c) |975 |

|40 |You meet a random person who is not employed. What is this individual most likely to be like? |

|a) |Someone who spends time outdoors instead of reading. |

|b) |Someone who is depressed and spends all day thinking about past mistakes. |

|c) |Someone who is slightly troubled by economic hardships. |

|Study Guide |

|Exam #4 |

|PSY 211 |

Disclaimer:

• This is a guide of the main points that I consider to be important. You are responsible for all material covered in the book or in lecture, unless otherwise noted by me.

Formulas:

• You will be given a formula sheet (attached) and the t-table from the book.

Not on this Exam:

• Do not use the formulas or d values in the book. Particularly, for the t-test, we used some simplified version of the formulas in class, which are easier, and still provide a close estimate of the longer formulas

• There are no story problems based on the one-tail t-test, you just need to know how it differs from a two-tail test.

Suggested Learning Strategies:

• Study in small groups

• Ask questions or e-mail Mike

• Know how to do all homework and study guide problems

• Make up your own multiple choice questions

Conceptual Questions:

• * Be able to interpret Output and work with results written in APA style (mainly t-tests, but correlation and regression Output are also fair game)

• * Be able to understand and define the differences between the z-test and the three types of t-tests. Given a story problem or list of data, be able to correctly identify which test to use.

• Know the four basic steps of hypothesis testing

• When writing out the null and alternative hypotheses, are we describing the expected difference in the sample or population?

• Know the difference between t-obtained and t-critical

• Know how probability relates to hypothesis testing

• How does the shape of the t-distribution differ from the shape of the z-distribution?

• How does sample size impact the critical t-value?

• What are degrees of freedom (df)?

• Why is it difficult to draw conclusions about group differences based on visual inspection of the data (such as, histograms)?

• What does a significant t-value mean?

• What does p = .000 really mean in the SPSS Output?

• Why is it possible for a result with a large effect size to be unreliable (not statistically significant)?

• Why is it possible for a small effect size to be reliable (statistically significant) in some cases?

• Know the main differences between the between-group and within-subject t-test

• Skepticism

Calculations:

• * Given r, r2, or d, be able to indicate the effect size

• Be able to calculate Cohen’s d, z-values, and various t-values given formulas

• Range of possible values for r, r2, and d

• If an obtained t-value is blacked out from SPSS Output, know how to calculate it from the remaining variables in the Output

Terms (may appear in multiple choice or short-answer):

• * p < .05

• Hypothesis testing

• Null hypothesis

• Alternative hypothesis

• Effect size

• Sampling error

• Iatrogenic

• Alpha (or significance) level

• Critical region

• Type I error

• Type II error

• Power

• One-tail vs. two-tail t-test (know the difference, but you won’t need to do any story problems based on one-tail tests)

• Pre-post study

• Manipulation check

• Matched sample

• Carryover effect

Formulas

I will also give you the t-table from the book!

You do not need to print them in advance to bring to the exam.

Beware, that in story problems, I might use “SD” as a symbol for standard deviation,

or “SE” to represent standard error of the mean.

Z = (M – μ) where σM = σ / [pic]

σM

Single sample t = (M – μ) where sM = s / [pic]

sM

Between-group t = [pic] where sM = [pic]

Cohen’s d for between-group t-test = [pic]

To find the value of s, average the standard deviation from each of the two subsamples, if available. If only the s for the combined sample is available, use that.

Within-subject t = MD / sMD where sMD = ([pic])

Cohen’s d for within-subject t-test d = MD / s

Old Exam Questions

I have taken these questions from last year’s exam. I think they will give you a good idea of the types of questions to expect, but ignore any that seem irrelevant! You can check answers here:

|3 |When writing the alternative hypothesis, researchers are describing differences that are expected to occur at this level |

|a) |sample |

|b) |construct |

|c) |population |

|4 |If SPSS shows that a correlation has a p-value of 0.12, the correlation is said to be |

|a) |statistically non-significant |

|b) |statistically significant |

|c) |not enough information |

|5 |If the alternative hypothesis is really correct, and based on sample data you reject the null hypothesis, you have made a |

|a) |Type I error |

|b) |Type II error |

|c) |correct decision |

|6 |A hypothesis that men and women do not differ on intelligence |

|a) |Null hypothesis |

|b) |Alternative Hypothesis |

|c) |One-tail hypothesis |

|8 |How will choosing a more strict alpha level impact the critical t-value? |

|a) |Increase it |

|b) |Decrease it |

|c) |No impact |

|9 |If Cohen’s d = - 4.39, the result will be significant |

|a) |Always |

|b) |Sometimes |

|c) |Never |

|10 |Type of t-test commonly used in twin studies |

|a) |single-sample |

|b) |between-group |

|c) |repeated-measures |

|12 |If the null hypothesis is really true, and your sample supports the alternative hypothesis, you have made a |

|a) |Type I error |

|b) |Type II error |

|c) |correct decision |

|13 |Effect size when Cohen’s d = .78 |

|a) |small |

|b) |medium |

|c) |large |

|14 |As sample size increases, the z value needed for significance will |

|a) |increase |

|b) |stay the same |

|c) |decrease |

|17 |Effect size when r2 = 0.26 |

|a) |small |

|b) |medium |

|c) |large |

|18 |Which of the following distributions is the flattest? |

|a) |z distribution |

|b) |t distribution where n = 5 |

|c) |t distribution where n = 10 |

|19 |The p-value shown in SPSS indicates the probability that |

|a) |the null hypothesis is true |

|b) |the alternative hypothesis is true |

|c) |the probability of a Type II error |

Below is the Output from a recent homework problem, in which you were to run two

t-tests, comparing gender differences in both “beauty concerns” and “war support,”

but some of the information has been removed. Use these charts for #21-22.

[pic]

[pic]

|21 |What is the value for the degrees of freedom? |

|a) |greater than 325 |

|b) |325 |

|c) |less than 325 |

|22 |What is the t-value for the “beauty concerns” t-test? |

|a) |greater than 2.37 |

|b) |2.37 |

|c) |less than 2.37 |

Below is some new Output, from our classroom data file, but not from any particular homework assignment. Instead of looking at gender differences, these t-tests examine differences across people who described themselves as an athlete (“Yes”) or not an athlete (“No”). I compared non-Athletes and Athletes along five variables: leadership, beauty concerns, physical complaints, authority problems, and life stress. These variables are very straightforward (no tricks), but I can answer questions if you want any details. Use these charts for #23-25 (next page).

[pic]

[pic]

|23 |How many of the five results shown are statistically significant? |

|a) |1 |

|b) |2 |

|c) |3 |

|24 |Based on these findings, if you met a random athlete, how might you expect them to differ from a non-athlete? |

|a) |More life stress |

|b) |More authority problems |

|c) |Fewer physical complaints |

|25 |What percentage of the sample self-reported that they were an athlete? |

|a) |less than 52.3% |

|b) |52.3% |

|c) |more than 52.3% |

|26 |Calculate a z-score, given the following information, and indicate whether it is | |

| |significant. | |

| | | |

| |Sample mean = 200 |z = |

| |Sample mode = 210 | |

| |Sample standard deviation = 160 | |

| |Population mean = 250 |Significant? yes/no |

| |Population mode = 248 | |

| |Population standard deviation = 111 | |

| |Sample size = 16 | |

| | | |

| | | |

| | | |

|28 |Psychologists have been fascinated by lottery winners. Using standard 9-point | |

| |happiness rating scales, they have tracked how their level of happiness changes after| |

| |winning. In fact, 1 year after winning the lottery, most winners (n = 9) note that | |

| |they are about 0.75 points happier (SD = 2.5). Calculate the within-subject t |t = |

| |statistic, Cohen’s d, and note whether the result is statistically significant. | |

| | |d = |

| | | |

| | | |

| | |Significant? yes/no |

More Old Exam Questions

Only review questions that appear relevant. Check out the answers online:



|1 |This statistic is not an example of an effect size |

|a) |p |

|b) |r |

|c) |d |

|2 |What is a hypothesis? |

|a) |Broad research question, aim, or goal motivating a particular study |

|b) |Detailed rationale explaining the logical basis for an observed finding |

|c) |Concise, testable statement describing the relationship between two variables |

|3 |Drug companies generally design drugs and then try to find a use for them at a later date. In one study, they examine whether a|

| |new drug impacts appetite. What type of hypothesis test should they use? |

|a) |No-tail test |

|b) |One-tail test |

|c) |Two-tail test |

|4 |What did William Kingdon Clifford say? |

|a) |“The danger to society is not merely that it should believe wrong things, though that is great enough; but that it should become|

| |credulous, and lose the habit of testing things and inquiring into them; for then it must sink back into savagery.” |

|b) |“But if we are empiricists, if we believe that no bell in us tolls to let us know for certain when truth is in our grasp, then |

| |it seems a piece of idle fantasticality to preach so solemnly our duty of waiting for the bell.” |

|c) |“It is better to be a human being dissatisfied than a pig satisfied, better to be Socrates dissatisfied than a fool satisfied. |

| |And if the fool or the pig are a different opinion, it is because they only know their own side of the question.” |

|5 |A null hypothesis states describes that there are no differences expected to occur at this level: |

|a) |measurement |

|b) |sample |

|c) |population |

|6 |Effect size for r2 = .26 |

|a) |small |

|b) |medium |

|c) |large |

|7 |The null hypothesis indicates: |

|a) |the effect will be negative, or opposite of expected |

|b) |an obtained difference will exceed chance |

|c) |no effect or no important difference |

|8 |Term indicating that results from any two particular studies might differ slightly due to chance: |

|a) |sampling error |

|b) |standard error of the mean |

|c) |carryover effects |

|9 |The results of a multiple regression indicate that R2 = .20, p = .50. Is this result statistically significant? |

|a) |non-significant |

|b) |significant |

|c) |not enough information |

|10 |Using a more strict alpha level for a z-test will cause the critical z-value to |

|a) |become more extreme |

|b) |stay the same |

|c) |become less extreme |

|11 |What type of test might you use to determine whether female doctors reliably differ from their romantic partners on IQ? |

|a) |correlation |

|b) |between-group t-test |

|c) |within-subject t-test |

|12 |What does a p-value describe? |

|a) |the probability that the result is reliable |

|b) |the probability that the result is due to chance |

|c) |the probability that the result will replicate |

|13 |Assume that the alternative hypothesis is accurate. If the researcher finds no effect, what has occurred? |

|a) |Type I error |

|b) |Type II error |

|c) |Iatrogenic error |

|14 |Effect size for Cohen’s d = .36 |

|a) |small |

|b) |medium |

|c) |large |

|15 |You have people rate how much they enjoy Coke and how much they enjoy Pepsi. The mean enjoyment rating for Coke is a bit higher|

| |than for Pepsi, and you find d = 1.25, t = 1.67, p = .63. How would you most likely get a result like this? |

|a) |large effect but a small sample |

|b) |unusual sample |

|c) |poor methodology or survey items |

|16 |This was not an example of a repeated-measured t-test discussed in class: |

|a) |Changes in tension before and after massage therapy |

|b) |Changes in physiological arousal before and after crossing an old bridge |

|c) |Changes in sleepwalking before and after sleep deprivation |

|17 |Which of the following distributions is the flattest? |

|a) |z distribution |

|b) |t distribution where n = 3 |

|c) |t distribution where n = 37 |

|18 |Effect size for r = .42 |

|a) |small |

|b) |medium |

|c) |large |

|19 |This type of test is not used to compare a treated sampled to an untreated population. |

|a) |z-test |

|b) |single-sample t-test |

|c) |between-group t-test |

|20 |There is no “depression gene” yet every year some researcher claims to have found it. The results are often reported on CNN and|

| |the other news networks, who fail to realize they are making this error. |

|a) |Type I error |

|b) |Type II error |

|c) |Hindsight bias |

|21 |Carefully examine the Output below for a series of between-group t-tests. The analyses compare “Early Birds” to “Night Owls” on|

| |five variables. Some information is missing or covered up. How many of the five results show a reliable group difference? |

| |[pic] |

| |[pic] |

|a) |1 |

|b) |2 |

|c) |3 |

|22 |Using the SPSS Output on the previous page, calculate the t-value for the group differences in Agreeableness, and choose the |

| |answer below closest to that t-value. |

|a) |1.3 |

|b) |1.7 |

|c) |2.1 |

|23 |Using the SPSS Output on the previous page, what would be the degrees of freedom for the analysis examining group differences in|

| |Neuroticism |

|a) |277 |

|b) |278 |

|c) |279 |

|24 |Study the following correlation table very carefully. The name of the first variable is blacked out. Based on the pattern of |

| |correlations, what might this variable be? |

| |[pic] |

|a) |depression |

|b) |ACT score |

|c) |happiness |

|25 |What do you suppose the value would be for the correlation in the “???” box? |

|a) |r = -.11 |

|b) |r =.02 |

|c) |r = .08 |

Part II. Calculations [10 points]

26. Z-test – 4 points

|Calculate a z-score, given the following information, and indicate whether | |

|it is significant. | |

| |Z = |

|Sample mean = 18 | |

|Sample mode = 22 | |

|Sample standard deviation = 10.75 |Significant? yes / no |

|Population mean = 20 | |

|Population mode = 16 | |

|Population standard deviation = 12.45 | |

|Sample size = 256 | |

| | |

27. t-test – 6 points

|28 people enroll in a clinical trail for a new schizophrenia medication. | |

|During the trial, ten patients drop out, and three more commit suicide. On| |

|a symptom rating scale, patients see an average reduction of 2.2 points (SD|t = |

|= 4.3) for their symptoms. Calculate the within-subject t-value. Indicate| |

|whether the result is significant. In a sentence or two, describe how you | |

|would interpret this finding. |Significant? yes / no |

| | |

| | |

| |Interpretation: |

| | |

| | |

Part III. Short Answer [10 points]

28. Pick One of the following:

a) List 5 types of inferential statistics and note when you would use each.

OR

b) List and briefly describe 5 major differences between the between-group and within-subject approaches.

OR

c) In a short paragraph, describe how inferential statistics are used in the social sciences. Make sure to mention the following terms: hypothesis testing, critical value, p-value, and effect size.

-----------------------

???

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download