Psychological Statistics Final Exam

If I am drawing observations out of a hat while sampling with replacement, the probability of drawing a certain outcome

remains constant across all trials

From a statistical standpoint, the hypothesis we are actually testing is called the ____hypothesis; we do this because…

null; it is easier to collect evidence disproving something then evidence proving something

According to the text, the researcher’s hypothesis is almost always that…

there will be a treatment effect

Sometimes we fail to reject the null hypothesis when we should reject it. This is referred to as

a Type II error

Why do we need statistics?

-People are generally bad at accurate estimations of probability

-It’s important to be able to judge the quality and accuracy of the statistical information we encounter in everyday life

-Our personal judgment tends to be too influenced by our expectations to be accurate

-People tend to see patterns and meaning in random events

Two events are said to be independent if

the occurrence of one has no effect on the probability of the occurrence of the other

When we reject the null hypothesis, we are saying that…

in a world where there’s nothing to find, results like ours don’t occur very often

According to your text, we tend to observe normally-distributed data so often in everyday life because…

factors that push scores up or down tend to cancel each other out, so more scores fall into the middle and fewer to the extremes

What is a good reason to convert data to z scores?

It allows you to analyze your data in terms of the normal distribution without calculating a whole new set of values for the normal distribution

As discussed in lecture, the multiplicative rule of probability indicates…

that the probability of two or more events occurring simultaneously will be lower than the probability of any of them occurring individually

What are measures of variability?

-the interquartile range

-the standard deviation

-the range

-the variance

What is a starting assumption underlying statistical analysis?

over repeated samples, data will be normally distributed

According to the text, we determine how strict we’re going to be about calling results significant by setting the…

alpha level

If you want to study the effect of hormonal changes in adolescent boys, your population would be

all adolescent males

The standardized score we use to calculate probability on the standard normal distribution is called the_______, and is measured in units of_________.

z-score; standard deviation

If we need a measure of variability that is easy to interpret in terms of the actual values in the data set, we should use the…

standard deviation

According to your text, which measure of central tendency is the only one where every individual score matters?

the mean

What’s the biggest difference between an interval scale and a ratio scale?

Values of 0 are meaningful in ratio scales, but aren’t meaningful in interval scales

What are examples of Type II error?

-Concluding that two different lab sections are doing equally well when one is doing much worse

-Finding no difference between two treatment approaches when one has much better outcomes

-Giving a terminally ill person a clean bill of health

-Overlooking a student with an exceptional IQ when recruiting for a gifted-and-talented program

According to lecture, what role does the standard normal distribution play in determining the likelihood of getting a particular score?

The area under the curve represents the frequency with which different values occur, and thus how likely they are to occur

The subset of a population from whom we collect data in order to understand the population is our…

sample

The chief disadvantage of the median, when compared to the mean, is that

it is less stable than the mean from sample to sample

Why do we square our deviation scores when we compute variance?

Because if we averaged the deviations without squaring them, we would always get a value of 0

a z score of 1.5 represents a score that is

One and a half standard deviations above the mean

In the text, the area on the normal distribution that marks the point that we will reject the null hypothesis is called the ______region

critical/rejection

The psychological phenomenon you are interested in studying is represented by the…

dependent variable

If you have one score that is dramatically higher or lower in value than the rest of the scores in your data set, that score is an ______, and will_____.

outlier; make your mean higher or lower than it should be

What would be best measured with an ordinal scale?

Finish time in a 50-yard dash

According to the text, what is one thing we can do to reduce the chances of getting a false positive result?

Set our alpha level at .05 or lower

What is the “rejection region”?

The area on the normal distribution that describes the point where we can reject the null hypothesis

Why does Type I error occur?

Because rejection of the null hypothesis is based on the low probability that we are wrong, not the absolute certainty that we are right

Looking down on a parking lot, if you see that 20% of the cars in the lot are blue, and 15% of the cars in the lot are red, then how would you estimate the probability that the next car to leave the parking lot would be red or blue?

add the two percentages together

According to your text, on the standard normal distribution, a little more than 68% of data fall _____from the mean

one standard deviation

The variable that represents conditions manipulated by the researcher is the ______variable.

independent

In a normally-distributed data set, which measure of central tendency is most likely to be affected by a single extreme score?

the mean

When we compute a score from our sample data, that score is a ______, and when the same score is computed from the entire population, that score is a ______

statistic; parameter

The measure of central tendency that is most useful for our purposes because it is generally stable from sample to sample is the

mean

What is the correct statement of the null hypothesis for the paired-samples t-test?

𝜇𝐷 = 0

A design where you measure something before and after a treatment has been applied is called a _____ design and would be tested with a ______t-test

pretest/posttest; paired-samples

An independent-sample study comparing two treatment conditions has 18 total degrees of freedom. If the two samples are the same size, how many participants are in each sample?

10

Subtracting one score from another in a pair of scores from each participant gives us a ________score

difference

As discussed in class, what is an advantage of paired-samples t-test?

It assesses an outcome while taking initial differences into account

An example of a study used for an independent-samples t-test

Evaluation the difference in self-esteem between student athletes and non-athlete students

In an independent-samples t-test, the null hypothesis is best thought of as

the mean of population 1 is equal to the mean of population 2

According to lecture, t-tests in general consist of…

dividing the difference between two values by the amount of variability in your data

Examples of studies used for a paired-samples t-test

-Evaluating changes in verbal skills between the ages of 2 and 3 for a sample of girls

-Evaluating the effectiveness of a pain reliever by measuring how much pain is reduced after taking the medication

-Evaluating average GRE scores from the first administration to a second administration for the same group of students

-Evaluating the effectiveness of a cholesterol medication by comparing cholesterol levels before and after taking the medication

If we know how many opportunities the data in our set have to vary before arriving at a specific value, we know our data’s

degrees of freedom

According to your text, the paired-samples t-test is a…

within-subjects test

In a paired-samples study where 30 students had their sophomore GPA compared to their freshman GPA, what would the degrees of freedom be?

29

According to the text, we determine whether or not we reject the null hypothesis in a t-test by comparing______to______.

out t statistic; a critical value for t

According to lecture, inferential tests in general examine either ______or_______

differences; relationships

According to the text, the use of standard deviation to estimate standard error means that…

the estimate of standard error will change from sample to sample

According to lecture, the right inferential test for your data is a test that…

will provide information that directly relates to your research question

The null hypothesis in a paired-samples t-test is…

that the mean difference score is equal to 0

Research questions that examine difference between groups are often tested using_____tests.

univariate

According to the text, why are one-sample t-tests not a typical way to test a hypothesis?

They require population-level information, and we generally don’t have population-level information

What is a statement of the null hypothesis in an independent-samples t-test?

𝐻0: 𝜇1 = 𝜇2

According to lecture, what is an important factor in which test you select?

Whether or not it provides information that fits your specific question

According to your text, a design where you compare data from two linked individuals in a paired-samples t-test is called a ______design

matched-subjects

According to lecture,_______are tests that divide the difference between two values by a measure of variability, and______are tests that divide one source of variability by another

t-tests; ANOVAs

According to lecture, what a potential limitation of t-tests?

We’re only able to make comparisons between two things at a time

An example of a study appropriate for a one-sample t-test

Comparing average SAT scores from a pilot test program to the national average

You’ve proposed a study where you’re going to measure academic performance (in the form of exam grades) before and after participation in a tutoring program. What test would be the best one for this study?

paired-samples t-test

According to lecture, the central measure of variability in univariate tests is…

variance

An example of a study appropriate for a paired-samples t-test

Comparing individual pain tolerance with and without the use of acupuncture needles

As discussed in class, what is the primary tradeoff between one and two-tailed t-tests?

One-tailed tests are more likely to return significant results, but can only do so in one direction; two-tailed tests allow for significant results in both directions, but are less likely to return significant results

As discussed in class, what is an advantage of t-tests over calculating probability using z scores?

We are less likely to overestimate the sample variance in a t-test than when using z scores

A researcher was interested in examining the different levels of aggression shown by college students and prisoners with life sentences. Aggression was measured by surveys given during one session. What analysis should the researcher perform on the mean aggression scores of the groups?

independent samples t-test

In a one-sample t-test, the degrees of freedom are represented by

N-1

A study using an independent-samples t-test has one sample with n=20 and a second sample with n=15, comparing an experimental treatment to a no-treatment control. What are the total degrees of freedom for this study?

33

A test that looks at relationships between multiple continuous variables would be a _______test

multivariate

In order to do an independent-samples t-test, you need…

a single continuous dependent variable and an independent variable with no more than two levels

Because an independent-samples t-test compares data from one group to data from another group, it is a ______test

between-subjects

If you are less interested in a specific outcome than you were in allowing for different possible outcomes in your research, you would propose a ______ hypothesis

non-directional

Analysis of variance makes a distinction between ______ and _______

treatment variability; error variability

A _______ test is one with a single dependent variable and one or more independent variables

univariate

In a one-way ANOVA, as meaningful variability increases and error variability decreases…

the value of F will increase

The basic formula for calculation of the F statistic in a one-way ANOVA is…

division of group variability by error variability

If you wanted to test the fuel consumption rate of cars by brand of car and year of model, which would be the best test to perform?

Factorial ANOVA

A repeated-measures ANOVA has at least

one DV measured multiple times

According to the text, what are the factors that affect statistical power?

-Your sample size

-The amount of variability in your data

-The size of the effect you’re trying to detect

-Your alpha level

As discussed in class, statistical power is ______ and is typically maximized by ______

the ability to detect a significant effect (assuming there is one); increasing sample size

According to the text, in a factorial ANOVA, each group in an independent variable is called a ______, and each unique combination of groups from all independent variables is called a _______

level; condition

What test would be best to run if you want to examine changes in GPA over four years of college?

repeated-measures ANOVA

What role does SSsubjects play in a repeated-measures ANOVA?

it allows us to examine the unique intersection of multiple factors influencing behavior

What test would be best to run if you want to examine the simultaneous effects of gender and college major on self-reported life-satisfaction?

factorial ANOVA

Power is an estimate of…

our ability to avoid Type II error

If the first level of your IV has a variance of 32, the second level of your IV has a variance of 32, and the third level of your IV has a variance of 45, what is true?

You do not have homogeneity of variance

A factorial design has at least

two IVs and one DV

In an ANOVA, error variance represents

variance that is not attributable to the difference between groups in our experiment

If you ran a factorial ANOVA on perceived trustworthiness of both attractive and unattractive men and women and the only significant result was a main effect of gender, what would you conclude?

that looks don’t affect trustworthiness, just whether the person is a man or a woman

One of the important assumptions we make in doing a one-way ANOVA is that

observations are independent

What test would be best to run if you want to compare average exam grades across three sections of PSY 101?

one-way ANOVA

If you’re running a factorial ANOVA with three independent variables, where the first IV has two levels, the second has three levels, and third has three levels, you have a _______ factorial design

2 × 3 × 3

In a one-way ANOVA, MSgroups measures the difference between group means, and MSerror measures variability within each group. If we rejected the null hypothesis, what would we expect to find?

That MSgroups is larger than MSerror

The term “effect size” refers to

the actual size of the mean or difference between means

A significant difference on one IV is called a _______, a significant difference on a unique combination of two or more IVs is called a _______

main effect; interaction

In a repeated-measures ANOVA study with 10 people tested at 6 different time points, what would the value of dftotal be?

59

A significant value of F for a one-way ANOVA means

there is at least one significant difference somewhere among the groups

In a one-way or factorial ANOVA, the variance associated with the difference between levels of an IV and the overall mean is called _______variance, and the variance associated with the difference between individual participants and the mean for their level of the IV is called _____ variance

between-groups; within-groups