PSYCH 210 FINAL EXAM

UW Madison Psych 210 Final, Dr. Gallimore

What type of test is used if the DV of a data set is nominal or ordinal (categorical)?

nonparametric test (such as chi squared)

When should you use goodness of fit chi squared?

if there is one variable to an established population parameter, then use a goodness-of-fit test to determine wether your sample data is a “good fit for” or matches the established parameter.

When should you use a chi squared test for independence?

if there are two variables that are being compared. used to determine wether the distribution data fort one variable is independent of the distribution the other variable.

What type of test should be used if the DV is interval or ratio?

parametric test (such as t-test or ANOVA)

What type of parametric test should you use if the IV are continuous?

correlation or regression

What type of parametric test should be used if the IV are catergorical?

z-test, t-test, or ANOVA

What type of test should be used if there are two groups?

z-test pr t-test

What type of test should be used if you are given a population SD and population mean?

z-test

What type of test should be used if you have the population mean but not population SD?

t-test

What type of test should you use if you do not have any population parameters, and are comparing two groups?

t-test

What type of test should be used if your participants each experience a different level of your IV?

independent samples t-test

What type of test should be used if your participants experience both levels of your IV?

dependent samples t-test

What type of test should you use if you do not have any population parameters and are comparing 3 or more groups?

ANOVA

What type of test should you use if o not have any population parameters and are comparing 3 or more groups, and there are 2 or more IVs?

Factorial ANOVA

What type of test should you use if o not have any population parameters and are comparing 3 or more groups, and there is only 1 IV, but your participants only experience one level?

between subjects ANOVA

What type of test should you use if o not have any population parameters and are comparing 3 or more groups, and there is only one IV, but the participants experience all levels of your IV?

repeated measures ANOVA

Is the mean, median, or mode influenced by outliers?

mean, because the mean is affected by the value of the #, but the median is affected by the quantity.

Alice is hoping to get the highest grad on her exam. Her professor uses a survey to see what measure of central tend. of the class wants him to use to calculate exam scores. She got a good score on her exam, and a good amount of her class got 100 percent. Which test would be the best to give her the highest grade possible?

mean (would create a positive skew)

How do you calculate variance?

you sum the squares of your data (find the mean of your data and subtract each score by the mean), then square the sums, finally divide by the total number of scores.

1) Sally wants to see if SSRIs are more effective than therapy for depression, in college students. She surveys 120 randomly selected students from UW Madison that meet the diagnostic criteria for depression. What are the 120 students from UW Madison in this study?

sample

Which of these examples is NOT an assumption of a normal distribution?

he graph has a tail that is longer on the right side then the right side

If you are conducting research that uses continuous variables, which type(s) of graph is appropriate?

bar graph, and/or histogram

What is in the denominator of a Z score when you are using the formula to calculate a Z-score?

σ

What is the numerator of the probability formula conceptually?

The number of favorable outcomes

If you wanted to use the unit normal table to see what percent people on exam 4 scored above 85, which column(s) would you use form the unit normal table to help find your answer?

Column A (lists the Z scores), and Column C (the tail)

What is the purpose of calculating standard error?

It tells you how much your sample scores vary from the mean

Which of these statements about hypotheses is incorrect?

If you get a test statistic that falls within the critical region, you accept your alternative hypothesis

What does β represent in statistics?

Probability of a type two error

During which step of the 7-step process of hypothesis testing do you compute the test statistic?

step 4

When would you use a t-test instead of a z-test?

When you do not know your sample population parameters

What type of test has these assumptions: homogeneity of variance, populations are normal (ish) destructions, and each value is sampled independently?

Independent samples t-test

What is in the numerator of an independent samples t-test?

M1-M2

If you calculate a test statistic of t=1.78, from a sample of n=41, for a one tailed test, significance level of .05, what should your decision about your hypotheses be?

reject null hypothesis

Mark is designing a test to see how anxiety levels improve for participants of his study before and after CBT therapy. In the study each participant experiences both conditions (before treatment, and after treatment). What type of statistical test listed below would be BEST for Mark to use?

dependent samples t-test

Which answer gives a correct example of a factor and level for an ANOVA?

Factor: chain restaurants, Level: chipotle

What is the biggest difference between Repeated-Measures ANOVA, and Between Subjects ANOVA?

Repeated measures removes individual differences

Cali is studying a group of 30 college students at UW Madison. She is hoping to see how their stress levels change throughout the year. She is doing so by testing each of the 30 student’s cortisol levels once every 2 months. What type of test is Cali most likely using?

Repeated-measures ANOVA

Marissa has conducted a study to see if walking to class or taking the bus impacts (Factor A) and setting one alarm or multiple alarms (Factor B) impact how many times students are late to class at UW Madison throughout the semester. She uses a two-factor ANOVA to determine if there are main effects or an interaction. She finds that There is a mean difference between walking to class and taking the bus, which suggests what?

There is a main effect for factor A

Maggie is taking Psych 210 right now, while working on an exam she chooses an answer that says, “correlation does establish causation,” does Maggie need to choose a new answer? If so, what part of the answer would need to be changed to make it correct?

Maggie did not choose the correct answer, there needs to be the word “not” added in front of establish to make the answer correct

Which of these are the characteristic of correlation?

strength, form, and direction

While taking Exam 4 in Psych 210 Kara was struggling to figure out what the correct order would be for which values are stronger. The list she was given was: -.73, .58, -.88, .30. What would be the correct order from least strength to most strength for these values?

.30, .58, -.72, -.88

What would be the correct way to describe the relationship between variables of a negative correlation between ft above sea level and temperature?

As the ft above sea level increases, the temperature decreases

Obsessive Compulsive Disorder (OCD) is a mental illness in which those effected by the illness have intrusive thoughts, and to lower anxiety (temporarily) they preform compulsions. For example, Avery is constantly obsessing over the thought of his dog dying, to ease these thoughts he switches the light on and off 5 times in his room every night before he goes to bed. Avery is convinced that these two things are correlated, however, even if he did switch the light on and off his dog could still die. This would be an example of what type of correlation?

Spurious correlation

If the data points on your graph are very spread out and when you draw a line around them it forms a circle like shape, what type of correlation is it most likely to be?

no correlation

A sample of favorite types of fruit includes apple, strawberry, kiwi, and grapes. This is an example of what level of measurement?

nominal

High and narrow is to __(    )__, as low and wide is to __(   )?__

leptokurtic, platykurtic

Interval & Ratio variables use these types of graphs EXCEPT...

pie chart

Why are Degrees of Freedom necessary?

To account for sample errors

What are the degrees of freedom for a population of N=20

 n=19

I have 70 erasers in a bag, 15 are pink, 20 are purple, 30 are yellow, and 5 are red.

 

6) What is the probability of picking a red eraser from the bag?

1/14

I have 70 erasers in a bag, 15 are pink, 20 are purple, 30 are yellow, and 5 are red. What is the probability of picking any color of eraser other than a purple?

50/70

A distribution has a m=200, s=20, *n*= 60. What is the shape of the distribution?

normal

If a sample has a mean of 50, standard deviation of 7 and 20 points are added to every score in the distribution what is the new mean and standard deviation?II

*µ*= 70, s=7

Attempting to raise the number of sales a clothing store increased the number of sales they promoted each month. The program has now been in place for 6 months, and the clothing store conducts a hypothesis test to see if increasing the number of sales increased the number of sales made. What number of tails does this hypothesis test have?

one tailed, specifies the direction.

 r^2= .40, what percentage explains the relationship between the X and Y variables?

40%

r=.30, what percent of variability in x is accounted for y?

first need to do r^2: .09, so 9% of the variability in x is accounted for by y.

n=25, s=15, calculate the estimated standard error of the mean

3

Confidence Intervals tells us what about the data?

specific direction of effect

A two tailed test with a t= 2.6 at an alpha of .05, how large does the sample size have to be in order to reject the null hypothesis?

6

In an independent-measures t-test two separate samples each with n=15 what are the degrees of freedom?

df=28

When rejecting a null hypothesis, the correct formatting for the p value is what?

p < .05

In a related samples T-test where n=25, SD=10, M1=2, M2=10, what is the confidence interval?

16 to 0

In a one-way ANOVA (between subjects) df between groups = 2, df within groups = 8,S S between groups = 80, F= 20. What is SS within groups?

SS within groups = 4

In a one-way ANOVA (between subjects) F(4, 50)=3.45, how many treatment conditions were compared?

5

In a one-way ANOVA (between subjects) F(4, 50)=3.45, how many participants were in the study?

55

For a One-Way ANOVA (repeated measures) test what is the effect size when SS between groups= 20, SS within groups=30 and SS between subjects=10?

0\.5

SS for factor A + SS for factor B + SS for the interaction between factor A and B sums to what?

SS between groups

Conceptually what are the degrees of freedom for Pearson’s correlation?

df regression, df residual

When SP= 14, SSx=2, My=5, Mx=6 what is the correct regression equation?

Ŷ= 7x-37

For a Chi Square goodness of fit test there are 4 levels, what are the degrees of freedom?

3

What is descriptive statistics?

procedures used to describe, organize, and summarize data. For example, central tendency, correlation, and regression

What is inferential statistics?

procedures that allow us to draw conclusions or make inferences that extend beyond the immediate data at hand. For example, hypothesis testing

Describe the nominal, ordinal, interval, and ratio scale

nominal consists of categories that have different names. Ordinal consists of a set of categories that are organized in sequence. Interval consists of ordered categories that are all intervals of exactly the same size. Ratio is an interval scale with the additional feature of an absolute zero.

What is a discrete variable? continuous variable?

A discrete variable consists of separate categories with apparent limits. A continuous variable has real limits, where you determine the unit of measurement and divide that in half. This is the cutoff of that variable.

describe a simple frequency distribution

Two categories; x and f. rank orders from highest at the top to lowest at the bottom. Must include all x-values, even if they have a frequency of 0

describe a grouped frequency distribution

groups made up into intervals consisting of the scores. There should be around 10 intervals, with the range being either 2, 5, 10, or 20.

what are real limits and when do we need to use them?

can be used in a simple or grouped FD. We need to use them when looking at continuous variables. The real limits are 1/2 a unit above, and 1/2 a unit below the actual numerical value

what is the difference between a bar graph and a histogram?

bar graph needs to be used with discrete data, and histogram with continuous data.

what is a percentile rank?

the percentage of individuals in the distribution with scores at or below the particular value

when should you use the median rather than the mean to describe central tendency?

With ordinal data, skewed distributions, open-ended distributions (ex:8+), or in a distribution with undetermined scores

what is a weighted mean and when should it be used?

It is the mean of two sets of scores, where n is different in each set

what happens to the mean when we add or subtract a constant from every raw score?

the same constant will be added or subtracted from the mean

what happens to the mean when we multiply or divide every raw score by a constant?

the mean will change in the same way

what are the four measures of variability?

1) range

when should you use variance?

Use when you want to take all scores into account, and to get a more representative idea of the spread

What is meant by the statement, the sample variance is a biased estimate of the population variance?

It tends to underestimate population variability. To fix this divide sample by n-1. Dividing by a smaller number provides a larger result and make sample variance an accurate and unbiased estimator of population variance

what are degrees of freedom? How do they play into correcting bias in sample variance?

n-1. The number of scores in a distribution whose values are free to vary, when the toady set of scores me a constraint.

what happens to the variance and standard deviation when we add or subtract a constant? multiply or divide?

add or subtract by a constant- no change

what is a z-score?

a descriptive statistics to standardize scores regardless of distributions. A z-score tells you how far away a particular score falls from the mean, in standard deviation units. The number tells you the distance between the score and the mean in terms of the number of standard deviations

what are the advantages of standardizing scores in a distribution?

We can accurately compare scores from different distributions

what is the mean and standard deviation of a standardized (z-score) distribution?

mean=0

how is probability used with the normal distribution?

a normal curve can be described by the proportions of area contained in each section. Probability=area under the curve

Does SIQ take variability between scores into account?

no

what does the process of random sampling do?

ensures that each individual within a population has an equal probability of being chosen for inclusion in a sample

What is a quasi experiment?

An experiment where the independent variable is a quasi-independent variable which means that it cannot be changed, for example breed of dog. Cause and effect relationships cannot be inferred with these experiments.

what is a deviation score?

a raw scores deviation from the mean of the distribution

what is sampling error?

the amount of discrepancy, or amount of error, between a sample statistic and it's corresponding population parameter?

what is the difference between a sample distribution and a sampling distribution?

a sample distribution is a distribution of raw scores (x). A sampling distribution is a distribution of statistics obtained by selecting all the possible samples of a specific size from a population. In a sampling distribution means tend to pile up around the population mean, the distribution is approx. normal, and in general the larger the n, the closer the sample means should be to the population mean.

What is the central limit theorem?

Gives a precise description of the distribution that would be obtained if you selected every possible sample, calculated every possible sample mean, and constructed the distribution of the sample mean. It would have a mean the same as the population mean, and will approach a normal distribution as n approaches infinity

describe the standard error

it is the standard deviation for a distribution of sample means. It provides a measure of how much distance is expected on average between a sample mean and the population mean