Advanced Psychological Statistics Final

You are working for a renowned head hunting company in midtown Manhattan. Your job is to identify high potentials for the financial services industry. The prior probability that someone you will consider is a high potential is 0.03. If someone is a high potential, their probability of having a degree from an Ivy League school is 0.6 and the probability that they have an Ivy League degree if they are not a high potential is 0.05. The person you are considering has an Ivy League degree. What is the probability that they are a high potential?

.27

You run an ANOVA model and want to determine eta squared. You know that the total sum of squares is 1250. The number of between group degrees of freedom is 4 and the mean square between is 100. Eta squared is?

.32

A general interpretation of the average is that of A. An expected value B. A measure of dispersion C. An unexpected value D. A measure of diversion E. A measure of unity

An expected value

What are primary applications of Bayes theorem? A. Updating of beliefs B. Optimal integration of prior and new information C. Determination whether someone is acting virtuously D. Both a) and b) E. Both a) and c)

Both a) and b) Updating of beliefs Optimal integration of prior and new information

Find the matching pairs A. Conditional probability B. Intersection C. Union 1. Addition rule 2. Multiplication rule 3. Lack of independence

Conditional Probability - Lack of independence Intersection - Multiplication rule Union - Addition rule

Conceptually, a correlation is A. The median normalized by the product of the standard deviations B. The covariance normalized by the sum of the standard deviations C. The covariance normalized by the product of the mean average deviations D. The covariance normalized by the product of the standard deviations E. The mean normalized by the product of the standard deviations

The covariance normalized by the product of the standard deviations

In the general linear model, "beta" refers to A. Type II error B. The weights of the predictors C. "Misses" D. The weights of the dependent variables E. The weights of the outcomes

The weights of the predictors

The probability of the intersection of two independent events A and B can be calculated as A. p(A) - p(B) B. p(A) + p(B) C. p(A) * p(B) D. p(A) / p(B) E. p(A)^p(B)

p(A) * p(B)

The standard error of the sample mean (SEM) is derived by taking the standard deviation of the sample and dividing it by A. n^2 B. n C. square root of n -1 D. n - 1 E. square root of n

square root of n

One in a hundred million people is a genius. If they are a genius, the probability that the person is acting extremely quirky is 0.999. If someone is not a genius, the probability that they are acting extremely quirky is 0.01. Someone is acting quirky - what is the probability that they are a genius?

.000001

You are working for an agency that is in charge of protecting the United States against an Ebola epidemic and wonder whether you should screen or quarantine passengers from certain countries. The probability that any given passenger has Ebola is one in a million. The probability that he is traveling from Liberia if he has Ebola is 0.9. Overall, one in 10,000 passengers is traveling from Liberia. What is the probability that a passenger has Ebola if he is traveling from Liberia?

.01

The probability that someone has blonde hair is 0.2. The probability that someone has an IQ over 115 is 0.15. Assuming that IQ and hair color are independent, what is the probability that someone picked at random is both blond and has an IQ over 115?

.03

You randomly sample 5 inmates in a maximum security prison. You note that all 5 exhibit levels of aggression that are higher than the median of the general population. The probability of obtaining this result by chance (if inmates in maximum security prisons do not differ in their levels of aggression from the general population) is

.03

The probability of being depressed is independent of having dark hair. The probability of being depressed is 0.1. The probability of having dark hair is 0.5. The probability that someone has dark hair and is depressed is:

.05

You compare the happiness of people forced to smile with those of people forced to frown. Both groups have the same number of participants. The mean happiness level in the smile group is 20. The mean happiness level in the frown group is 19. The standard deviation of both groups is 20. The effect size is

.05

You study whether playing action video games increases eye hand coordination. Given your data, the probability of the data given the null hypothesis is 0.06, and the probability of the data given the alternative hypothesis is 0.95. The Bayes factor is

.06

This Bayes Factor constitutes the strongest evidence against the null hypothesis A. 0.5 B. 0.3 C. None of them do, as they fail p < 0.05 D. 0.4 E. 0.1

.1

You are working in a primary care office. Flu season is starting. For the sake of public health, it is critical to diagnose people with the flu. The prior probability that someone who is walking through your door has the flu is 0.1. If someone has the flu, their probability of having a runny nose is 0.99. However, if someone doesn't have the flu, e.g. just has the cold (they will (think that they) have something, otherwise they wouldn't seek out your office), their probability of having a runny nose is 0.9. Someone comes in and has a runny nose. What is the probability that this person has the flu?

.11

The probability that someone is drunk at a party is 0.7. The probability that someone is getting into a fight at a party is 0.1. The probability that someone is drunk and getting into a fight is 0.08. What is the probability that someone is getting into a fight if they are drunk?

.114

You study whether childhood adversity has an impact of stress levels as an adult. Given your data, the probability of the data given the null hypothesis is 0.1, and the probability of the data given the alternative hypothesis is 0.85. The Bayes factor is

.117

Imagine you have an ANOVA table with the following values: Corrected Model - Sum of Squares: 1205 Intercept - Sum of Squares: 3500 Error - Sum of Squares: 4900 Corrected Total - Sum of Squares: 6105 Total - Sum of Squares: 9605 What proportion of the variance can be accounted for by the model underlying the ANOVA? A. 0.3 B. 0.125 C. 0.05 D. 0.2 E. 0.635

.2

You are looking for a great person to spend your life with. The prior probability that someone is a great match is 0.1 (and I'm being generous here). If the person is a great match, their probability of having a good job is 0.8, whereas 30% of people who are not a great match also have a good job. On the first date you learn that the person has a good job. What is the probability that they are a great match?

.23

You do a multi-factorial ANOVA and your eta squared is 0.25. This means that your model explains what percent of the observed variance in the dependent variable?

.25

You randomly sample 4 attendants of a power posing seminar. You note that 3 of the 4 exhibit higher levels of confidence than the median of the general population. The probability of obtaining this result by chance (if those attending power posing seminars do not differ in their levels of confidence from the general population) is

.25

The first date went well (.23). You are still dating the same person and still looking for a great person to spend your life with. You are cautiously moving forward with the relationship and know that if they are a great match, their likelihood of being kind to animals is 0.9 whereas their likelihood of being kind to animals if they are not a great match is 0.7. You see that your prospective spouse is kind to animals. What is the probability that they are a great match now?

.28

You are interested in oddness research. You read a paper that claims that study participants who were assigned an odd ID number will more readily admit to reading the horoscope. The rationale is that activating the concept of "oddness" leads people to behave in nontraditional way. The study reports a significant difference in horoscope reading between the group of participants with odd and even IDs. The alpha level is 0.05, power was 0.2 and - given that this is all a bit far-fetched - the ratio of true to false effects in oddness research is 0.1. What is the probability that the reported effect is real? A. 0.2 B. 0.05 C. 0.29 D. 0.1 E. 0.15

.29

You roll 10 dice and note the sum of this first roll. You keep the results of 3 dice, but roll the other 7 again, then noting the sum (of all 10 dice) of this second roll. You do this (throwing the dice in this way) 100 times. The expected correlation between the sum of the first and the sum of the second rolls is:

.3

25% of addicts are using both cocaine and heroin. 70% of addicts use cocaine. What is the probability that an addict is using heroin if they are using cocaine?

.35

You study whether feeling lonely makes you anthropomorphize inanimate objects. Given your data, the probability of the data given the null hypothesis is 0.2, and the probability of the data given the alternative hypothesis is 0.5. The Bayes factor is

.4

You are interested in the psychological implications of high SES. You read a paper that claims that higher social class people behave more unethically. The paper reports a significant difference in the propensity of high and low SES people behaving unethically. The significance level of this study was 0.05, the ratio of true to false effects in this field is 0.35, but the author was anxious to issue a press release and prepare his TED talk, so power is low, at 0.1. The probability that this study reports a real effect is A. 0.95 B. 0.41 C. 0.35 D. 0.58 E. 0.1

.41

The probability that someone's last name starts with a vowel is 0.4. The probability that someone's last name starts with the letter "S" is 0.1. What is the probability that someone's last name either starts with a vowel or the letter S?

.5

You want to do a psychological experiment. On any given day, there is a 30% chance that your participant won't show up. In addition, there is a 10% chance that the stimulus presentation software won't work. There is also a 5% chance that there is a problem with saving the data and a 3% chance that there are other problems like experimenter error. For the sake of simplicity, you can assume that all these events are independent. What is the probability that you will be able to do a successful experiment (be able to record useful data) on any given day?

.58

The prior probability that an entering undergraduate student will get their BA is 0.48. 65% of students who graduate entered with a good high-school GPA. 40% of entering undergraduate students who don't graduate also had a good high-school GPA upon entering. The probability of an entering college student getting their BA if they have a good high school GPA is

.6

You are still interested in oddness research and you read a similar - but not the same - paper on the effects of oddness priming on people doing odd things. The paper reports a significant effect of oddness priming. Alpha is also 0.05 and R of the oddness research field is still 0.1, but this was a highly powered study, with power = 0.95. What is the probability that the effect reported in this study is real?

.655

From empirical studies, we know that the lifetime prevalence for depression is 10%. We also know that the probability of someone being both female and depressed at the same time is 0.07. What is the probability that someone who is depressed is female?

.7

The prior probability that an entering undergraduate student will get their BA if they have a good high school GPA is given by the answer of the previous question (.6). In addition, 90% of students who graduate with a BA also have a high SAT score, whereas 55% of students who drop out have high SAT scores. The probability that an entering freshman who has a good high-school GPA and high SAT scores will graduate with a BA is

.71

The probability that someone is cheating on their spouse is 0.3. The probability that someone cheats on their exam if they are cheating on their spouse is 0.9. The probability that someone cheats on their exam if they are not cheating on their spouse is 0.1. The probability that someone is cheating on their spouse if they are cheating on the exam is

.79

40% of prisoners have tattoos and get into fights with other inmates. 50% of prisoners get into fights with other inmates. What is the probability that a prisoner that is getting into a fight has a tattoo?

.8

The probability that someone is drunk at a party is 0.7. The probability that someone is getting into a fight at a party is 0.1. The probability that someone is drunk and getting into a fight is 0.08. What is the probability that someone was drunk if they got into a fight?

.8

You study whether self-affirmation activates the right hemisphere of the brain. Given your data, the probability of the data given the null hypothesis is 0.4, and the probability of the data given the alternative hypothesis is 0.5. The Bayes factor is

.8

You are still interested in the psychological implications of high SES. You read a different paper from the same field that looks at a similar, but not the same phenomenon. The alpha level in this study was 0.05, R in this field is still 0.35, but this researcher was much more careful, with power of this study at 0.9. The paper reports a significant effect. What is the probability that it is real?

.86

You are in college and picked a couple of classes for this semester. You wonder which classes you should drop, if any. 25% of classes suck and should be dropped. If a class sucks, the professor is kind of an idiot 70% of the time. Regardless of whether their classes suck or not, 20% of professors are kind of idiots. You are in a class and observe that the professor is kind of an idiot. What is the posterior probability that the class is going to suck (and ought to be dropped)?

.88

Instead of power posing in horses, you are interested in whether music can induce a mere exposure effect at short timescales. You read a paper that claims that it can, reporting a significant mere exposure effect. Alpha level was 0.05, the ratio of true to false effects in this field is 0.7 and the power of this study was 0.8. The probability that this effect is real is :

.92

You are part of a profiling unit for serious crimes. The probability that a serial killer lives in the neighborhood is 1 in a million. The probability that there are multiple murders in the neighborhood if there is a serial killer is 0.99. The lifetime probability of there being one murder in the neighborhood is one in a thousand and the probability of three murders in a neighborhood within a year is 1 in a million. What is the probability that there is a serial killer on the loose if there were three murders within a year in the neighborhood?

.99

An ideologue is certain that his position is correct. New evidence is emerging that challenges this position. There is only a 0.01% chance of such evidence emerging if the position is correct. However, there is an 95% chance of such evidence emerging if the position is not correct. Now, after such evidence has emerged, if the ideologue is Bayesian, how should this new evidence impact his position - how certain to be correct (expressed as a posterior probability) is he now?

1

You do a psychophysical experiment on the consistency of appreciation of aesthetic beauty by humans. You pick 2 observers and ask each of them to make 100 aesthetic judgments of a painting. When estimating mean and variability of these judgments, you have this many degrees of freedom A. 0 B. 199 C. 100 D. 200 E. 1

1

You compare the IQ on people on NZT with people not on NZT. Both groups have the same number of participants. The mean IQ in the NZT group is 125, the mean IQ in the control group is 100. The standard deviation of both groups is 15. The effect size is

1.66

You compare the number of depressive episodes in a control group vs. a group of participants on antidepressants. Both groups have the same number of participants. The mean number of depressive episodes in group 1 is 10. Then mean number of depressive episodes in group 2 is 5. The standard deviation of group 1 is 3. The standard deviation of group 2 is 2. The effect size is

1.96

If there are two candidates running in a political primary and they receive an equal number of votes in a county, the outcome is sometimes decided by a coin toss. What is the chance that one candidate would win all coin tosses, if there are a total of 6 of them - assuming that the coins are fair?

1/64 (1.5%)

Expected Frequency Observed Frequency Fresh&Co 50 18 Pret A Manger 50 82 In the above example, the number of participants (n) is ______ whereas the number of groups (k) is ______: A. 2; 100 B. 100; 2 C. Cannot be determined D. 2; 50 E. 50; 2

100; 2

Finally, an estimate of the stability of the sample mean of 100 and with a standard deviation of 10 but based on 100 people in the sample yields of 95% confidence interval with an upper bound of? A lower bound of?

101.96 98.04

You are also interested in the 90% CI for the same data. The critical z-value for such a CI is 1.645. Therefore, the upper limit of the CI is? The lower bound is?

103.678 96.32

You want to calculate how precisely your methods estimate a sample mean. In order to do this, you want to calculate the confidence interval. The sample mean is 100 with a standard deviation of 10. The critical z-value to cut off a 95% confidence interval is 1.96 and there are 20 people in your sample. The upper bound of the confidence interval is? The lower bound of the CI for the same data is?

104.38 95.62

Now you want to calculate the 99% confidence interval for the same data. The critical z value for a 99% CI is 2.58. Therefore, the upper bound of the 99% CI is? The lower bound of the CI for the same data is?

105.77 94.23

The total number of all possible interaction effects in a full general linear model for a 5x5x5x5 ANOVA is A. 3 B. 4 C. 375 D. 5 E. 11

11

What is the number of t-tests you would have to do if you wanted to test all possible mean differences between 16 groups? A. 1 B. 120 C. 2 D. 105 E. 16

120

Imagine you are doing a one-way ANOVA and have the following table: Between groups sum of squares: 1905 Total sum of squares: 3200 What is the within groups sum of squares? A. 500 B. 1295 C. 1.68 D. 5105 E. It can't be determined from the information given. To answer this question, we would also need the number of levels and the degrees of freedom.

1295

You are interested in the effects of sleep deprivation on self regulation. You take a group of 150 people and measure how many cookies they eat while you interview them once when sleep-deprived and once when fully rested. You do a paired-samples t-test to analyze the data. You have how many degrees of freedom?

149

You present Nature with the numbers 3, 6 and 9 and ask whether they are members of the set. Each time, Nature answers that yes, these numbers are members of the set. You conclude that the rule is: "Consecutive multiples of 3 are members of the set". Which of these is the best (maximally informative) number to try next in order to test this rule? A. 12 B. 21 C. 18 D. 2 E. 15

2

You wonder if weather (rain, cloudy or sunny) makes a difference on mood (happy or not). To determine this, you sample the moods of 200 people on a sunny day, 200 people on a rainy day and 200 people on a cloudy day and record whether they were happy or not. You then want to do a chi square test on this data. What are the degrees of freedom? A. 200 B. 599 C. 199 D. 3 E. 2

2

The F-distribution has this many parameters A. 4 B. 3 C. 0 D. 2 E. 1

2 (degrees of freedom between and within)

How many parameters are needed to completely determine a given normal distribution? A. 3 - The mean, the standard deviation and the skewness B. 2 - The mean and the standard deviation C. 1 - The mean D. 4 - The mean, the standard deviation, the mean average deviation and skewness E. 0 - There is only one

2- the mean and the standard deviation

You decide to study the effects of smoking, drinking and partying on life satisfaction. To do so, you assign people randomly to one of two smoking conditions (smoking or not), one of three drinking conditions (no alcohol, 1 drink per day, several drinks per day) and partying conditions (no partying, 1 hour of partying per day, 2 hours of partying per day). This design has A. 3 factors, 18 total conditions B. 1 factor, 18 total conditions C. 3 factors, 12 total conditions D. 2 factors, 18 total conditions E. 3 factors, 24 total conditions

3 factors, 18 total conditions

You are interested in whether NYU students prefer Fresh&Co or Pret A Manger, so you give 100 NYU students a choice between a gift card to either place and record their choices. You then conduct a chi square test: Expected Frequency Observed Frequency Fresh&Co 50 18 Pret A Manger 50 82 What is the chi square value? A. 3.53 B. 52.4 C. 92.5 D. 18.0 E. 40.8

40.8

How many 4-way interaction effects are there in a 2x2x2x2x2 ANOVA? (Hint: Make a table or draw this) A. 16 B. 4 C. 5 D. 32 E. 2

5

You are ambitious and want to do a 9 factor ANOVA. You would expect the full model to have how many terms?

512

You do a PCA (principal component analysis) on a dataset from 256 EEG electrodes recorded when a participant was meditating. You can treat the data from each electrode as a variable. The PCA returns 256 factors. The 1st factor has an eigenvalue of 70, the 2nd factor an eigenvalue of 40, the 3rd an eigenvalue of 20, the 4th an eigenvalue of 10, the 5th an eigenvalue of 7, whereas the 6th and 7 have eigenvalues of 5 and 3, respectively. The eigenvalues of the remaining factors are under 1. Using the Kaiser criterion, how many factors should you consider as "real"? A. 7 B. 1 C. 3 D. 256 E. 5

7

An independent samples t-test with two samples and 41 observations in the first sample and 39 observations in the second sample has the following number of degrees of freedom A. 39 B. 78 C. 79 D. 80 E. 41

78

You have 100 observations in a sample, and you calculate 4 parameters like the sample mean from this sample. How many "degrees of freedom" do you have left for further calculations? A. 100 B. 99 C. 94 D. 96 E. 95

96

You want to know whether hand washing makes people feel morally superior. You look at all published papers on this effect and make a histogram of the p-values reported in the literature. The following distribution of p-values would be most indicative of p-hacking and publication bias A. A bimodal distribution with one of the peaks just below p = 0.05 B. An exponential distribution C. A uniform distribution D. A cauchy distribution E. A bimodal distribution with one of the peaks just below p = 0.95

A bimodal distribution with one of the peaks just below p = 0.05

Induction is A. A method to sample from non-normal distributions. B. A method to make conclusions about specifics from general principles. C. A method to derive general principles from specific instances. D. A different term for voltage. E. A method to create certainty.

A method to derive general principles from specific instances.

Deduction is A. A method to deal with uncertainty. B. A method to derive general principles from specific instances. C. A method to make conclusions about specifics from general principles. D. A different term for voltage. E. A method to sample from non-normal distributions.

A method to make conclusions about specifics from general principles.

You want to do a 5-way 5x5x2x2x2 ANOVA. Assuming that there is no effect of any of the treatments, if you compared all groups with t-tests, at an alpha-level of 0.05, you would expect about this many false positives (type I errors) A. About 1000 B. About 10 C. About 200 D. About 1 E. About 5

About 1,000

A t-test for independent groups makes the following assumptions A. The measures in the population from which we sample are distributed normally B. The populations we compare have equal variance C. The measures we take are at least on interval-scale level D. The observations are independent of each other E. All of the above

All of the above

Given measurements on a ratio scale, we can A. Meaningfully interpret the distance between two numbers B. Meaningfully interpret the ratio between two numbers C. Meaningfully interpret the relative magnitude of two numbers D. Assume that the scale has an absolute zero E. All of the above

All of the above

In a binary world with only two possible and mutually exclusive outcomes A and B, the probability of B can be arrived at by calculating A. p(B) B. 1-p(A) C. p(~A) D. All of the above E. None of the above

All of the above

What is true about degrees of freedom? A. They denote the number of values in a calculation that are free to vary B. They need to be adjusted when estimating a sample parameter from the sample itself and when using other parameter to do so that were also estimated from the sample (e.g. sample standard deviation that utilizes the sample mean). C. To avoid "double-dipping", we have to reduce the degrees of freedom for each parameter we calculate from the sample itself. D. All of the above E. None of the above

All of the above

Why are positive (significant) results easier to publish, in principle? A. Much easier to interpret B. Much less common - if one compares everything with everything, most things won't be meaningfully related C. Inability to find something might reflect low power on behalf of the researchers D. Inability to find something might reflect inadequate operationalization of IV or DV, not anything about reality. E. All of the above

All of the above

Why is the MBTI so unreliable? A. Because the underlying distribution of scores in each dimension is unimodal, not bimodal. B. Because there are 4 independent dimensions and we categorize each one. C. Because the scores of most people fall close to the decision boundary in each dimension. D. All of the above E. Because the professor doesn't like Jung

All of the above

Why does correlation not imply causality? A. Correlation is just a statistical relation B. The relation between two variables in a correlation is bidirectional C. There are other variables that could mediate the observed correlation D. All of the above E. This is a trick question - correlation does imply causality

All of the above Correlation is just a statistical relation The relation between two variables in a correlation is bidirectional There are other variables that could mediate the observed correlation

You want to measure how aggressive someone is. But first, you need to operationalize "aggressiveness". In order to do so, you could A. Ask the person how aggressive they are, on a scale from 1 to 5 (1 = not at all, 3 = sort of, 5 = very much so) B. Ask someone who knows the person to rate them how aggressive they are, on a scale from 1 to 5 (1 = not at all, 3 = sort of, 5 = very much so) C. Review their criminal record in order to determine how many prior convictions for violent crimes they have D. All of the above are - a priori - equally plausible operationalizations of "aggressiveness". E. None of the above are plausible operationalizations of aggressiveness

All of the above are - a priori - equally plausible ope rationalizations of "aggressiveness"

The basic idea behind an ANOVA is to A. Compare expected and unexpected frequencies B. Compare expected and observed frequencies C. Analyze variance by dividing it in a known and an unknown part D. Analyze variance by dividing it in variance between and within groups E. Analyze variance by dividing it in a reliable and an unreliable part

Analyze variance by dividing it in variance between and within groups

Mean squared deviations A. Are derived by dividing the sum of squared deviations by the degrees of freedom. Doing this separately between and within groups yields a ratio that corresponds to an F value. B. Are derived by dividing the sum of squared deviations by the degrees of freedom. Doing this separately between and within groups yields a product that corresponds to an F value. C. Are meaningless. They should use median squares instead. D. Are a synonym for sum of squares. E. Are derived by dividing the degrees of freedom by the sum of squared deviations. Doing this separately between and within groups yields a ratio that corresponds to an F value.

Are derived by dividing the sum of squared deviations by the degrees of freedom. Doing this separately between and within groups yields a ratio that corresponds to an F value.

What is the basic idea behind the Mann-Whitney U test? A. Square the differences between expected and observed differences and sum them up. B. Arrange all values from both distributions in order, then assign ranks and calculate the sum of ranks from one sample, then compare it with the ranksum of the other sample C. Plot the cumulative probability distributions of the two samples, then find all differences between the two distributions and sum them up D. Plot the cumulative probability distributions of the two samples, then find the largest separation between the two distributions E. Take the absolute mean difference between the two samples and divide it by the pooled standard deviation

Arrange all values from both distributions in order, then assign ranks and calculate the sum of ranks from one sample, then compare it with the ranksum of the other sample

Given measures on an ordinal scale, it is meaningful to interpret the A. Mode B. Median C. Mean D. Both a) and b) E. Coefficient of variation

Both a) and b) Mode and median

Why can't we use the sum of simple differences from the sample mean as a dispersion measure? A. This measure would always be zero. B. This measure is meaningless, as the positive and negative deviations from the sample mean cancel out, due to how the sample mean is defined. C. Both a) and b) D. Squaring makes sure that the numbers are integers E. Taking the absolute value makes sure that we don't have to deal with negative numbers, and who needs that kind of negativity in their life?

Both a) and b) This measure would always be zero. This measure is meaningless, as the positive and negative deviations from the sample mean cancel out, due to how the sample mean is defined.

The t-test is designed to establish differences between sample means for these situations A. Unknown population variation B. Small sample size C. Large sample size D. Known population variation E. Both a) and b)

Both a) and b) Unknown population variation Small sample size (but you should always have a large sample size)

Why is it not a good idea to run a t-test if one has ordinal data that is not normally distributed A. A t-test compares means. The mean is not meaningful on the level of ordinal data B. A t-test compares medians. So one needs higher level data than that. C. A t-test compares means. The mean is only a good representation of the distribution if it is normal D. Both a) and c) E. This is a trick question. It is always a good idea to do a t-test.

Both a) and c) A t-test compares means. The mean is not meaningful on the level of ordinal data A t-test compares means. The mean is only a good representation of the distribution if it is normal

It is most suitable to do a logistic regression when A. Predicting binary outcomes B. Predicting continuous outcomes C. When predictors and outcomes are linked in a sigmoidal way D. When predictors and outcomes are linked in a linear way E. Both A and C F. Both A and D G. Both B and D

Both a) and c) Predicting binary outcomes When predictors and outcomes are linked in a sigmoidal way

When would it be better to use a permutation test instead of a t test? A. When the distribution of the mean differences is not normal B. When the distribution of the mean differences is not exponential C. When the distribution of the mean differences is not known D. When the distribution of the mean differences is not t E. Both A) and C)

Both a) and c) When the distribution of the mean differences is not normal When the distribution of the mean differences is not known

An event A is independent of an event B if A. The probability of A equals the probability of A given B B. The probability of A and B happening must equal zero C. The probability of the intersection of A and B is equal to the probability of A times the probability of B D. Both a) and c) E. Both b) and c)

Both a) and c): The probability of A equals the probability of A given B The probability of the intersection of A and B is equal to the probability of A times the probability of B

What are some advantages and concerns associated with the Bonferroni correction for multiple comparisons? A. It brings the overall alpha level back to the intended level, but increases the risk of missing real effects for any given comparison. B. It increases the power of the test, but inflates the alpha level. C. It increases the power of the test, but will lead to more false positives. D. It helps to reduce the number of false positives, but reduces power. E. Both a) and d)

Both a) and d) It brings the overall alpha level back to the intended level, but increases the risk of missing real effects for any given comparison. It helps to reduce the number of false positives, but reduces power.

Given measurements on a nominal scale, numbers should be interpreted as A. Labels B. Means C. Products D. Categories E. Both a) and d)

Both a) and d) Labels and categories

Regression to the mean is likely to be a problem in A. Cases with a high correlation between predictor and outcome B. Cases with a low correlation between predictor and outcome C. Repeated measures designs with relatively low reliability D. Both A and C E. Both B and C

Both b) and c) Cases with a low correlation between predictor and outcome Repeated measures designs with relatively low reliability

Everything else being equal, smaller samples (relative to larger samples): A. Are better estimators of population parameters B. Are more variable from each other C. Have larger standard deviations D. Are less representative of the population E. Both B) and D)

Both b) and d) Are more variable from each other Are less representative of the population

What are data? A. Data are given to you - by gods and/or nature (depending on your metaphysics) B. Data are made by people C. Data are both "born and made" D. Data come about as a result of a measurement process E. Both c) and d)

Both c) and d) Data are both "born and made" Data come about as a result of a measurement process

Computer science is the study of algorithms to automate processes. A computer scientist specializes in the theory of computation and the design of computational systems. Given this definition, is computer science a science? A. Yes, it says it right in the name. B. Yes, it is really hard and the people studying it like to do a lot of math. C. No. Computer science is generally concerned with abstract statements about mathematical structures. D. No. There is no data in computer science, generally speaking E. Both c) and d)

Both c) and d) No. Computer science is generally concerned with abstract statements about mathematical structures. No. There is no data in computer science, generally speaking

The black swan problem concerns A. The uncertainty of deduction. B. The plural of anecdotes. C. The fact that induction can never delivery certainty D. The fact that something that has never happened before is not impossible E. Both c) and d)

Both c) and d) The fact that induction can never delivery certainty The fact that something that has never happened before is not impossible

When deriving posterior probabilities with Bayes theorem, their magnitude critically depends on A. The likelihood differential B. Reliability C. The prior probability D. Your personal beliefs E. Both the prior probability and the likelihood differential

Both the prior probability and the likelihood differential

Someone asserts that "Frogs are green". How would you test this assertion? A. By finding a green frog B. By reading all the books on frogs you can get your hands on and see what colors frogs are listed as C. By remembering that all the frogs you have ever seen were green D. By asking the world authority on frogs what color frogs are E. By looking for a frog that is not green, e.g. a red frog.

By looking for a frog that is not green, e.g. a red frog.

You want to develop a drug to increase IQ. So far, you have created 4 candidate substances - A, B, C and D. They all shifted the group IQ mean (tested on 30 volunteers each) away from the population mean. You calculated the following parameters A: z-score of 2.5 B: Group mean = 115, SD = 15 C: Group mean = 130, SEM = 2 D: Group mean = 130, SEM = 4 Which of the 4 outcomes is most unlikely - and thus most promising to increase IQ?

C

Find the matching pairs A. Certainty B. Uncertainty C. Data 1. Induction 2. Deduction 3. Science

Certainty - Deduction Uncertainty - Induction Data - Science

You do a study on whether an antidepressant drug is effective and show that there is a significant difference between people who do and who don't get the drugs in terms of their depressive symptoms. In reality, the drug does not work. You A. Made a correct rejection B. Did a great job and should expect a promotion. C. Committed a type II error D. Missed a real effect E. Committed a type I error

Committed a type 1 error

You want to develop a new drug that improves creativity. Despite your best efforts, you cannot show that there is a significant difference between the creativity scores of people who do vs. don't take the drug. In reality, the drug is actually effective in improving creativity. You A. Correctly rejected a false hypothesis B. Will have an easy time publishing these results C. Committed an alpha error D. Committed a type I error E. Committed a type II error

Committed a type II error

The basic idea behind a Chi-square test is to A. Analyze variance by dividing it in a known and an unknown part B. Analyze variance by dividing it in variance between and within groups C. Compare expected and observed frequencies D. Analyze variance by dividing it in a reliable and an unreliable part E. Compare expected and unexpected frequencies

Compare expected and observed frequencies

Both reliability and validity are - at their core - based on A. Experiments B. Independent variables C. Ordinal scales D. Dependent variables E. Correlations

Correlations

Imagine you do an ANOVA and plot the means. On the y-axis is the group mean, on the x-axis different levels of one factor. Different levels of another factor are plotted as separate lines. What would you expect these lines to look like if there were only an interaction effect, but no main effects? A. Crossing B. Different slopes C. Horizontal D. Vertical E. Parallel

Crossing

Find the most appropriately matching pairs A. Data B. Knowledge C. Causality 1. Science 2. Experiment 3. Measurement

Data - Measurement Knowledge - Science Causality - Experiment

You develop a new drug for the treatment of depression. You give this drug to 30 volunteers but withhold it from 30 others (you give them a sugar pill instead). You then measure the difference in depression symptoms before and after via the Beck Depression Inventory (BDI) in both groups. You calculate that the probability of obtaining the observed difference in mean BDI scores between groups is 0.06. Given this situation, the most reasonable course of action is to A. Do a new study with a larger number of participants to discern whether the drug has a modest effect that was unlikely to be detected, given the small sample size B. Conclude that the drug does not work - 30 participants in each group is already more than enough, as the central limit theorem applies at that point. C. Conclude that the drug does not work - Fisher has spoken. D. Conclude that the drug works - 0.06 is close enough. What's a difference in 0.01 between friends and who is Fisher anyway? E. Conclude that the drug does not work - there is nothing to be done. Such is life.

Do a new study with a larger number of participants to discern whether the drug has a modest effect that was unlikely to be detected, given the small sample size

Showing that your experimental result replicates (holds) in a population you haven't studied yet is a good example of A. Face validity B. Construct validity C. Reliability D. Internal validity E. External validity

External validity

The t-distribution is similar to the normal distribution, but at smaller n, it has A. A broader peak B. Fatter tails C. Skew D. Thinner tails E. A taller peak

Fatter tails

You work for a pharmaceutical company on a new anti-anxiety drug. You do an ANOVA to test for side-effects. The ANOVA is significant, so there will be side-effects. Your CEO asks you to do the most liberal post-hoc test available to find conditions that are likely to cause side-effects (and avoid lawsuits in case you miss some serious - but rare - side effects). You recommend to do the following kind of post-hoc test: A. Fisher's least significant differences (LSD) test B. A Scheffe test C. A Newman-Keuls test D. Tukey's honestly significant differences test (HSD) E. A Dunnet test

Fisher's least significant differences (LSD) test

A researcher wants to know the effects of presenting participants with randomly placed dots on a screen on reaction time. To study this, each participant is exposed to one condition, either 5, 10 or 15 randomly placed dots. This is an example of a A. Repeated measures ANOVA B. Mixed effect factor C. design with 5 levels D. Fixed effect factor E. Random effect factor

Fixed effect factor

In principle, this would make a null-result much more easy to publish - as it is more informative A. Lower power B. Higher power C. p < 0.05 (assuming alpha = 0.05) D. This is a trick question, they are never meaningful. E. Higher variance

Higher power

Which statement about instruments of measurement is (most) accurate? A. If an instrument is not objective, it cannot be reliable or valid B. If an instrument is not reliable, it cannot be objective or valid C. If an instrument is not valid, it cannot be objective or reliable D. If an instrument is not objective, it cannot be reliable or accurate E. If an instrument is not valid, it cannot be objective or accurate

If an instrument is not objective, it cannot be reliable or valid

Which statement about instruments of measurement is accurate? A. If the instrument is not reliable, it can't be valid B. If the instrument is not valid, it can't be objective C. If the instrument is not valid, it can't be objective D. If the instrument is not reliable, it can't be objective E. If the instrument is not valid, it can't be reliable

If the instrument is not reliable, it can't be valid

In practice, increasing power is usually achieved by A. Increasing the mean difference B. Increasing sample size C. Increasing effect size D. Increasing the alpha-level E. Increasing the standard deviation

Increasing sample size

You come down with flu-like symptoms. The doctor on call does a test for a lethal and incurable tropical disease (think Ebola in 2014). You have not been to the tropics recently, and in this population, 1 in a billion people have this disease. The manufacturer of the test reports that if you do have the disease, it will come back positive 99.9999% of the time. The test sheet also reports that the false positive rate is only 1 in a million. The test comes back positive. The doctor says this means that you have the disease beyond a reasonable doubt. What should you do? A. Insist on a re-test B. Thank the doctor and congratulate them on the quality of statistical education in medical school. C. Give up on life D. Sue the manufacturer for disseminating false information. E. Put your affairs in order

Insist on a re-test

Imagine that the figure below represents group means of a 2x2 ANOVA (factors A and B). B1 and B2 are crossing, B2 is higher at point A1 and B1 is higher at point A2. The most likely interpretation is that A. Interaction and A are significant, but B is not. B. Both A and B are significant, but the interaction is not C. Both interaction and B are significant, but A is not significant D. A is significant, B is not significant, and the interaction is also not significant E. Both A and B are not significant, but the interaction is

Interaction and A are significant, but B is not.

Standardized tests of achievement like the SAT are a good example of measures on a A. Interval scale B. Ordinal scale C. Fish scale D. Nominal scale E. Ratio scale

Interval scale

A key characteristic of science is that it: A. Is hard. B. Involves induction. C. Creates undeniable facts. D. Involves deduction. E. Involves both induction and deduction in an iterative fashion.

Involves both induction and deduction in an iterative fashion

The mean absolute deviation (MAD) A. Is a less robust measure of dispersion than the standard deviation. B. Is a more robust measure of central tendency than the standard deviation. C. Is a less robust measure of central tendency than the standard deviation. D. Is a more robust measure of dispersion than the standard deviation. E. Is a more robust measure of dispersion than the median.

Is a more robust measure of dispersion than the standard deviation

The principal problem of the range as a measure of dispersion is that it A. Is hard to calculate B. Requires ratio scale data C. Is extremely sensitive to outliers D. Is correlated with the mean E. Is too robust to outliers

Is extremely sensitive to outliers

All of this is true about eta squared, except for A. It can be calculated by dividing the Sum of squares between by the Sum of squares total B. It is a measure that is analogous to R squared in correlation and regression. C. It can be calculated by dividing the Sum of squares between by the Sum of squares within D. It can be calculated by dividing the product of F times the between degrees of freedom by the product of F times the between degrees of freedom plus the within degrees of freedom. E. It is a measure of the variance explained by the ANOVA

It can be calculated by dividing the Sum of squares between by the Sum of squares within

The probability of the intersection of A and B is equal to A. The probability of A times the probability of B B. The probability of B alone C. The probability of A alone D. The probability of the union of A and B E. It can't be determined from this information, it depends on whether A and B are independent or not

It can't be determined from this information, it depends on whether A and B are independent or not

You are interested in whether one's undergraduate major affects entry-level salary (which you know is not normally distributed). To this end, you randomly assign NYU freshmen to major in: psychology, philosophy, math, and German. Which nonparametric test should you conduct to test the hypothesis that college major has an impact on entry level salaries? A. Kruskal-Wallis test B. Kolmogorov-Smirnov test C. Mann-Whitney U test D. Wilcoxon rank-sum test E. Wallisch-Cachia test

Kruskal-Wallis test (equivalent of nonparametric ANOVA)

Everything else being equal, decreasing the likelihood differential will make the posterior A. It sets it to 1 B. This is a trick question. What actually changes is the prior. C. Its probability doesn't change D. Less likely E. More likely

Less likely

You measure IQ in a group of 200 people with severe learning disorders. In general, this sample is described well by a mean of 75 and a standard deviation of 15. However, one person who absent-mindedly wandered in from the Mensa convention next door was also tested and measured at an IQ of 142. The measure of central tendency that is most affected by this is likely the A. Mean B. Median C. Mode D. Standard Deviation E. Mean average deviation

Mean

The Wilcoxon rank-sum test compares this parameter between two samples A. This is a trick question. It compares nothing of the sort. It is a *non*-parametric test, remember? B. Modes C. Peaks D. Medians E. Means

Medians

Everything else being equal, increasing the probability of the prior will make the posterior A. It makes it zero B. Less likely C. The likelihood doesn't change D. More likely E. This is a trick question. Changing the probability of the prior changes the likelihood differential.

More likely

Is medicine a science A. No, it is primarily concerned with fixing things that are wrong - specifically with healing the sick - not with a principled understanding of the natural world. B. Yes, it is quite rigorous C. Yes, it is very useful D. Yes, doctors use tests to make decisions E. Yes, medicine is awesome

No, it is primarily concerned with fixing things that are wrong - specifically with healing the sick - not with a principled understanding of the natural world.

Is math a science? A. No. It doesn't even say so. All real sciences have a "science" in the name, e.g. neuroscience. B. Yes, math is inductive, so why wouldn't it be a science? C. Yes. It is really, really hard. D. Yes, of course. It is not only a science, it is the queen of all the sciences. E. No. Math is entirely deductive, so it doesn't qualify as a science.

No. Math is entirely deductive, so it doesn't qualify as a science

Taking note of someone's gender is a good example of taking measures on a A. Interval scale B. Fish scale C. Ratio scale D. Ordinal scale E. Nominal scale

Nominal scale

As the number of observations in a sample increases, the distribution of the sample means will increasingly approximate which distribution, regardless of the population distribution we sample from, if the central limit theorem applies? A. F distribution B. Uniform distribution C. Gamma distribution D. Abnormal distribution E. Normal distribution

Normal Distribution

The t-distribution approximates this distribution with larger n A. Uniform distribution B. Abnormal distribution C. F distribution D. Gamma distribution E. Normal distribution

Normal distribution

Multiple linear regression is indicated when A. One is doing an experiment B. One has a single predictor C. One has multiple predictors D. One has binary outcomes E. Both B and D

One has multiple predictors

Asking someone how much pain they are in, on a scale from 0 to 10 is a good example of a A. Ratio scale B. Fish scale C. Ordinal scale D. Interval scale E. Nominal scale

Ordinal scale

Imagine you do an ANOVA and plot the means. On the y-axis is the group mean, on the x-axis different levels of one factor. Different levels of another factor are plotted as separate lines. What would you expect these lines to look like if there were only main effects, no interaction effects? A. Vertical B. Crossing C. Horizontal D. Parallel E. Different slopes

Parallel

You are a grand strategist considering a military operation. The success of plan A relies on the following assumptions to be true (which you can assume to be independent): That the enemy doesn't spot you first, which you estimate as 0.9, that there will be artillery support, which you estimate at 0.9, that there will be air support, which you estimate at 0.9, that there will be sufficient fuel for your tanks, which you estimate at 0.9, that all elements of your combined arms effort are coordinating well, which you estimate at 0.9 and that the weather cooperates, which you also estimate at 0.9. Or plan B, which relies on a daring paratrooper drop on the enemy HQ, the success of which you estimate at 0.65. Assuming your estimates are accurate, which of these two plans should you adopt in order to maximize the probability of success of the operation?

Plan B

What are the most suitable pairs? A. Plane B. Line C. "S" 1. Simple linear regression 2. Multiple linear regression 3. Logistic regression

Plane - Multiple linear regression Line - Simple linear regression S - Logisitc regression

What is the basic idea behind the Kolmogorov-Smirnov test? A. Take the absolute mean difference between the two samples and divide it by the pooled standard deviation B. Plot the cumulative probability distributions of the two samples, then find the largest separation between the two distributions C. Square the differences between expected and observed differences and sum them up. D. Plot the cumulative probability distributions of the two samples, then find all differences between the two distributions and sum them up E. Arrange all values from both distributions in order, then assign ranks and calculate the sum of ranks from one sample

Plot the cumulative probability distributions of the two samples, then find the largest separation between the two distributions

This is considered a good measure to combat publication bias A. Insisting on reporting confidence intervals B. Insisting on reporting effect sizes C. Meta-analysis D. Pre-registration E. Insisting on reporting p-values

Pre-registration

If events A and B are not mutually exclusive, their joint probability (A happening or B happening) is given by the A. Probability of A and B minus the probability of A or B B. Union C. Probability of A plus the probability of B plus the probability of A and B D. Intersection E. Probability of A plus the probability of B minus the probability of A and B

Probability of A plus the probability of B minus the probability of A and B

Doing many studies, but publishing only the ones that yielded a significant result is a good example of A. A legitimate research practice B. p-hacking C. Data falsification D. Publication bias E. Data fabrication

Publication bias

You want to know whether people who have beards are more perceived to be more trustworthy. As a student researcher, you only have funds to recruit 10 models. The most powerful design to reveal that there is a difference is to A. This is a trick question. 10 models are not enough to show anything about anything. B. Randomly pick 10 bearded models, ask people to rate them before and after the beard was shaved off then do a z-test. C. Randomly pick 10 bearded models, ask people to rate them before and after the beard was shaved off then do a t-test for correlated groups. D. Randomly pick 5 models with beards and 5 without beards, ask people to rate them, then do a z-test. E. Randomly pick 5 models with beards and 5 without beards, ask people to rate them, then do a t-test for independent groups.

Randomly pick 10 bearded models, ask people to rate them before and after the beard was shaved off then do a t-test for correlated groups.

You measure reaction times in a face recognition task. Most of your measures follow a normal distribution with a mean of 1.2 seconds and a standard deviation of 0.3 seconds. However, in one of the trials, the study participant didn't pay attention and didn't make a response for 20 seconds. The dispersion measure most affected by this is the A. Mean Average Deviation B. Median C. Range D. Standard deviation E. Mode

Range

Find the matching pairs A. Range B. Standard deviation C. Mean absolute deviation 1. Mean 2. Median 3. Mode

Range - Mode Standard Deviation - Mean Mean Absolute Deviation - Median

Reaction times are a good example of measures used in psychology that can be considered to be on a A. Ratio scale B. Interval scale C. Nominal scale D. Fish scale E. Ordinal scale

Ratio scale

Interval scales allow us to meaningfully interpret measures in all ways except for A. Their identity (whether they are the same or different) B. Their magnitude (if one is bigger than another) C. Ratios between two measures D. Differences between measures E. The mean of the measures

Ratios between two measures

Recording data of many dependent variables, testing all possible relationships and then reporting only those that are significant is a good example of A. Data fabrication B. p-hacking C. A legitimate research practice D. Data falsification E. Researcher degrees of freedom

Researcher degrees of freedom

All of these are examples of post-hoc tests except for A. SBK B. LSD C. SNK D. REGWQ E. HSD

SBK

Which of the following would be a plausible way to write the full model for a 2x3x4 ANOVA? A. Score = Grand mean + Effect Factor 1 + Effect Factor 2 + Effect Factor 3 + Interaction between Factors 1&2 + Interaction between Factors 1&3 + Interaction Interaction between Factors 2&3 + Error B. Score = Grand mean + Effect Factor 1 + Effect Factor 2 + Effect Factor 3 + Interaction between Factors 1&2 + Interaction between Factors 1&3 + Interaction Interaction between Factors 2&3 + Interaction between Factors 1&2&3 C. Score = Grand mean + Effect Factor 1 + Effect Factor 2 + Effect Factor 3 + Interaction between Factors 1&2 + Interaction between Factors 1&3 + Interaction Interaction between Factors 2&3 + Interaction between Factors 1&2&3 + Error D. Score = Effect Factor 1 + Effect Factor 2 + Effect Factor 3 + Interaction between Factors 1&2 + Interaction between Factors 1&3 + Interaction Interaction between Factors 2&3 + Interaction between Factors 1&2&3 + Error E. Score = Grand mean * Effect Factor 1 * Effect Factor 2 * Effect Factor 3 * Interaction between Factors 1&2 * Interaction between Factors 1&3 * Interaction Interaction between Factors 2&3 * Interaction between Factors 1&2&3 + Error

Score = Grand mean + Effect Factor 1 + Effect Factor 2 + Effect Factor 3 + Interaction between Factors 1&2 + Interaction between Factors 1&3 + Interaction Interaction between Factors 2&3 + Interaction between Factors 1&2&3 + Error

The two operations that underlie the calculation of the mean and the median, respectively are A. Summing and multiplying B. Ordering and summing C. Modulus and summing D. Summing and ordering E. Multiplying and ordering

Summing and ordering

Permutation tests give a better estimate of the true p value except when A. The assumptions about the underlying population distribution are met B. The assumptions about the underlying population distribution are violated C. The mean difference is high D. Variance is low E. The sample data it is based on is representative of the population of possible data

The assumptions about the underlying population distribution are met

The fundamental problem with induction is that A. The conclusions one draws with inductive methods can be wrong. B. It is the complement to deduction. C. One needs data to do it. D. It is only useful if one has a small number of observations. E. It is hard to do.

The conclusions one draws with inductive methods can be wrong

What is an effect size, conceptually? A. The mean divided by the standard deviation B. The mean difference divided by the pooled standard error C. The mean difference divided by the pooled mean average deviation D. The mean difference multiplied by the pooled standard deviation E. The mean difference divided by the pooled standard deviation

The mean difference divided by the pooled standard deviation

What is true about mean and median? A. The median is a more robust measure than the mean because it is less affected by outliers. B. The median is calculated by taking the average of the mean C. The mean is a more robust measure than the median because it is less affected by outliers. D. Both median and mean are equally robust measures of central tendency E. Both median and mean are equally robust measures of dispersion

The median is a more robust measure than the mean because it is less affected by outliers.

A given empirical F value (say 3) is less likely to occur by chance as A. The number of experimental groups increases B. The number of between group degrees of freedom increases C. The number of between group degrees of freedom increases and the number of within group degrees of freedom decreases D. The number of within group degrees of freedom increases E. The number of within group degrees of freedom decreases

The number of within group degrees of freedom increases

As the number of degrees of freedom increases, the chi-square distribution behaves as follows A. The peak of the probability distribution shrinks and moves to the right as the distribution broadens B. The peak of the probability distribution grows higher and moves to the right as the distribution narrows C. The peak of the probability distribution grows higher and moves to the right as the distribution broadens D. The peak of the probability distribution grows higher and moves to the left as the distribution broadens E. The peak of the probability distribution shrinks and moves to the left as the distribution broadens

The peak of the probability distribution shrinks and moves to the right as the distribution broadens

If events A and B are mutually exclusive, the probability of their union is equal to A. The probability of A added to the probability of B B. Their intersection C. The probability of A alone D. The probability of B alone E. The probability of the union minus the intersection

The probability of A added to the probability of B

What generative process - in nature - yields normal distributions? A. The random combination of a few independent factors B. The random combination of many independent factors C. The systematic combination of some factors D. The random combination of many dependent factors E. The random combination of a few dependent factors

The random combination of many independent factors

What does simple linear regression minimize? A. The sum of the cubed differences between predicted values and measurements B. The sum of the absolute values of the differences between predicted values and measurements C. It minimizes the regression to the mean D. The sum of the differences between predicted values and measurements E. The sum of the squared differences between predicted values and measurements

The sum of the squared differences between predicted values and measurements

A measurement is considered to be *not* reliable if A. They systematically depend on who makes them B. They are not standardized C. They don't correlate with other measures we expect from our theory D. They don't predict anything in the real world E. There is no consistency between repeated measures

There is no consistency between repeated measures

You want to know whether people who meditate are more mindful than those who don't. You do a t-test for independent samples and find that the t value is 1.5. The critical t-value given the effect size and degrees of freedom is 1.98. You conclude that A. There is a significant difference between the group means and the null hypothesis should be rejected B. the data is inconclusive. C. There is no significant difference between the group means and the null hypothesis should be accepted D. There is no significant difference between the group means and we fail to reject the null hypothesis E. You did the wrong test. A t-test for paired samples would have been more appropriate

There is no significant difference between the group means and we fail to reject the null hypothesis

A measurement is considered to have no criterion validity if A. They systematically depend on who makes them B. They don't correlate with other measures we expect from our theory C. They are not standardized D. There is no consistency between repeated measures E. They don't predict anything in the real world

They don't predict anything in the real world

A measurement is considered to be *not* objective if A. They are not standardized B. They don't predict anything in the real world C. They don't correlate with other measures we expect from our theory D. There is no consistency between repeated measures E. They systematically depend on who makes them

They systematically depend on who makes them

Find the matching pairs - doing x leads or corresponds to which y? A. Verifying B. Measuring C. Falsifying 1. Turkey problem 2. Science 3. Data

Verifying - Turkey Problem Measuring - Data Falsifying - Science

You are interested in whether sleep affects productivity, so you randomly assign people to one of 10 groups. Participants in group 1 sleep for 0 hours, participants in group 2 for 1 hours, participants in group 3 for 2 hours and so on. You record whether people are high performers. There are 100 high performers in total, and you want to see how they distribute amongst the groups. In version 1 of this experiment, 5 of the groups have 9 high performers and 5 have 11 high performers. In version 2 of this experiment, 1 group has 5 high performers, 1 group has 15 high performers and the other 8 groups have 10 high performer. Imagine doing a chi square test on both of these hypothetical outcomes. A. Since the scale of measurement is ordinal, it makes no sense to calculate chi squared values. B. Version 2 of the experiment will yield a larger chi squared value than version 1. C. As the sum of the deviations from the expected values is the same in both versions, both versions of the experiment will yield the same chi squared value. D. There is not enough information to determine whether both versions of the experiment will yield the same chi squared value. E. Version 1 of the experiment will yield a larger chi squared value than version 2.

Version 2 of the experiment will yield a larger chi squared value than version 1.

In class, we learned that PCA is based on a correlation matrix whereas MDS is based on a matrix of distances. Could one use a correlation matrix to do an MDS? A. Yes, why not? B. Yes, but only if one squares them first C. Yes, but one would have to subtract all values from 1, as 1 indicates closeness and 0 extreme distance, in terms of correlation. D. Yes, but only if there is no negative correlation present, as distances can't be negative E. No, this makes no sense

Yes, but only if there is no negative correlation present, as distances can't be negative

You have a significant ANOVA. Your professor tells you to do a conservative post-hoc test. You tell him that A. You will do a LSD test B. This is a bad idea. Just because the ANOVA is significant doesn't mean you won't have a multiple comparisons problem C. You will do LSD D. You will do a HSD test E. You will do all possible t-tests between the groups

You will do a HSD test

You are getting married tomorrow, at an outdoor ceremony in the arid steppe. In recent years, it has rained there only 5 days out of each year. Unfortunately, the forecasters have predicted rain for tomorrow. When it actually rains, they correctly forecast rain 80% of the time. When it doesn't rain, they still forecasted rain 20% of the time. The probability that it will rain on your wedding day is

about .03 - .07

As you increase the sample size, the standard error of the means (SEM) A. decreases as a function of the square root of the sample size B. decreases as a function of the sample size C. neither increases nor decreases D. increases E. This is a trick question. It is the population standard deviation that decreases.

decreases as a function of the square root of the sample size

Find the matching pairs A. p(A and B) = p(A) * p(B) B. Inverting simple probabilities C. Posterior of 1 D. Inverting conditional probabilities E. p(A+B) = p(A) + p(B) 1. Bayes theorem 2. 1 - p 3. Prior of 1 4. Statistical independence 5. Two mutually exclusive events

p(A and B) = p(A) * p(B) - Statistical independence Inverting simple probabilities - 1-p Posterior of 1 - Prior of 1 Inverting conditional probabilities - Bayes theorem p(A+B) = p(A) + p(B) - Two mutually exclusive events

Updating the prior probability and applying Bayes will allow you to compute the _______________ probability. Hint: "Conditional" is technically correct, but we are looking for a more specific kind of conditional probability here.

posterior