PSYC 210 Exam 2
Which isn't a type of hypothesis? A. Null hypothesis B. Alternative hypothesis C. Flipside hypothesis
C. flipside hypothesis
What does standard deviation indicate in a set of data? A. how reliable the mean is B. that the mean accurately reflects the population C. how much the data differs from the mean on a average D. the data is significantly different than a control set
C. how much the data differs from the mean on average
In the formula for the T-value, what does the denominator represent A. Standard Deviation B. Variance C. How much they differ by chance D. The difference between two sample means
C. how much they differ by chance
Which of the following does not apply to the standard normal distribution: A. The area under the curve is equal to 1.0 B. Probabilities range from 0 to 1.0 C. Makes judgements about the sample D. Makes judgements about the population
C. makes judgements about the sample
Which of the following distributions creates a distribution mean? A. Population B. Sample C. Sampling D. Correlation
C. sampling
Which of the following is NOT a type of t-test? A. T-test for independent groups B. T-test for correlations C. T-test for dependent groups D. T-test for correlated groups
C. t-test for dependent groups
In the equation, ŷ=bx+a, which one of the variables represents slope? X B A ŷ
B
True or False: Standard deviation is an indication of the reliability of the mean A. True B. False
B. false
What value is the Type I Error set to? .01 .05 .10 .15
.05
How many degrees of freedom do you lose for each mean in your test? 0 1 2 3
1
How many degrees of freedom is in a t-test for two independent samples of 15 people? 28 30 29 31
28
How many degrees of freedom in a t-test should be used when two samples are composed of 15 and 20 data points? 35 14 19 33
33
In a sample of 5 people, where we calculated the mean, how many degrees of freedom would we have? 6 5 4 3
4
Which of these Z scores would result in rejecting the null hypothesis of a two-tailed test? A. 2.00 B. 1.95 C. 1.75 D. -1.01
A. 2.00
A type 1 error is __________. A. A false positive B. A false negative
A. a false positive
What is the difference between a sample distribution and a sampling distribution? A. A sampling distribution is a distribution of different samples, a sample distribution is a sample of individual scores B. A sample distribution is normally distributed whereas the sampling distribution might not be C. A sample distribution has a smaller sample size D. There is no difference
A. a sampling distribution is a distribution of different samples, a sample distribution is a sample of individual scores
If you keep sample, the distribution will _____________. A. Become normal B. Not become normal
A. become normal
What statement is true about two types of t's? A. Correlated groups t - no individual difference variance B. Correlated groups t - individual difference variance C. Independent groups t - deals with neither variance between groups nor variance within groups D. Independent groups t - deals with variance between groups but not with variance within groups
A. correlated groups t- no individual difference variance
What is the requirement for something to be considered a random sample? A. Each member of the population has an equal chance of being chosen B. The sample method is randomly chosen C. There is an implicit bias towards a certain group
A. each member of the population has an equal chance of being chosen
What is an example of a situation in which we would use a correlated groups t-test? A. Examining results before and after a treatment in an individual B. Comparing grades in PSYC 210 across academic classes C. Comparing SAT scores between males and females D. Deciding if UNC is a better school than Duke
A. examining results before and after a treatment in an individual
What should you assume between two samples when performing a two-sample t-test? A. Homogeneity of variances B. Homogeneity of sample size C. Homogeneity of mean D. Homogeneity of significance
A. homogeneity of variances
What is the difference between a formal statement for a 2-tailed test vs a 1-tailed test? A. If it was 1-tailed, it would include a > or < sign. B. If it was 1-tailed, it would only use a < sign C. If it was 2-tailed, it would only use a < sign D. If it was 2-tailed, it would only use a > sign
A. if it is 1-tailed, it would include a > or < sign
What is the difference between correlated and independent groups for t-tests? A. Independent groups are unrelated, while correlated groups are related B. Independent groups are related, while correlated groups are related C. Correlated groups are bigger, while independent groups are smaller D. Correlated groups are smaller, while independent groups are bigger
A. independent groups are unrelated, while correlated groups are related
What is the standard error? A. Indicator of how reliable the mean is. B. Indicator of how far the data differs from the mean on average. C. Standard deviation of the sampling distribution. D. Tells a researcher how spread out the responses are
A. indicator of how reliable the mean is
What is true about the sample mean? A. It is used in a t-test. B. It is used in a z-test. C. Both are correct. D. Both are incorrect
A. it is used in a t-test
Inferential statistics is based on: A. Random sampling B. Total population C. Z-scores D. t-tests
A. random sampling
What is the difference between sampling distribution and population/sample distribution? A. Use standard error B. Use standard deviation C. It is one-tailed D. Use random sample
A. random sampling
Which of the following scenarios would be best to use a t-test for? A. Relationship between gender or race and income B. Comparing GPAs of two statistics classes at UNC C. Finding the correlation between age and reading level D. All of the above
A. relationship between gender or race and income
What is the difference between sample distribution and a sampling distribution? A. Sample distribution is a distribution of scores in a particular sample, and sampling distribution is a distribution of sample statistics (usually the mean). B. Sample distribution is a distribution of sample statistics (usually the mean), and sampling distribution is a distribution of scores in a particular sample. C. Sample distribution is a random sample, and sampling distribution is a non-randomly selected sample D. None of the above
A. sample distribution is a distribution of scored in a particular sample, and sampling distribution is a distribution of sample statistics (usually the mean)
What is the difference between standard deviance of the mean and standard error? A. Standard deviation is an indicator of how far data differ from the mean (on average), and standard error is an indicator of how reliable the mean is. B. Standard deviation is an indicator of how reliable the mean is, and standard error is an indicator of how far data differ from the mean (on average). C. They're a different phrase for the exact same thing D. None of the above
A. standard deviation is an indicator of how far data differ from the mean (on average), and standard error is an indicator of how reliable the mean is
Which one is a universal assumption? A. There is a homogeneity of variables B. The mean differences will not represent a normal distribution C. Both are correct D. None are correct
A. there is homogeneity of variables
True or False: In relation to the Central Limit Theorem, samples become more normal when N is very large A. True B. False
A. true
What kind of error results in rejecting the null hypothesis when it's true? A. Type I B. Type II C. Neither
A. type I
When do we lose a degree of freedom? A. When we calculate a statistic B. Only when we calculate the mean C. When we do not have the population standard deviation D. When we have a small sample size
A. when we calculate a statistic
When can you use a t-test? When we do not have the population standard deviation To test if two groups are different enough To test if a correlation is significant All of the Above
All of the above
What does r2 represent? Slope of a regression line Relative strength of a relationship Amount of variance in Y that can be explained by variance in X
Amount of variance in Y that can be explained by variance in X
Why do we use sampling distributions? A. To see correlation B. To make judgements about the population C. To see randomness of sample D. To find an outcome
B. to make judgements about the population
What does a t-test compare? A. Two population means B. Two sample means C. A population mean to a sample mean D. Two population means and two sample means
B. two sample means
What describes the phenomenon that as N (sample size) gets bigger, the sampling distribution becomes more normal? A. Centered Mean B. Standard Error C. Central Limit Theorem D. Statistical Power
C. central limit theorem
When would you use a one-tailed test over a two-tailed test? A. How SAT scores differ between states B. Are SAT scores higher in certain cities C. Does the amount of time watching Netflix differ between age groups D. Does the amount of drinking differ between ethnicities
B. Are SAT scores higher in certain cities
Which of the following statements correctly identifies the three types of distributions A. Populations distributions are actual distributions of part of the population often described by roman numerals, sample distributions are distributions of the entire population and sampling distributions is a distribution of sample statistics usually the standard deviation. B. Population distribution are actual distribution scores of the entire population often described with Greek letters, sample distribution is the distribution of scores in a particular sample, and the sampling is a distribution of sample statistics which is usually the mean.
B. Population distribution are actual distribution scores of the entire population often described with Greek letters, sample distribution is the distribution of scores in a particular sample, and the sampling is a distribution of sample statistics which is usually the mean.
What does one-sample t-test compare? A. A sample mean to a sample population B. A sample mean to a population mean C. A population mean to a sample population D. A sample mean to a sample population
B. a sample mean to a population mean
Which of the following is NOT an assumption for a t-test? A. Mean difference is sampled from a normal distribution of mean differences B. Both samples were random and are dependent on one another C. Variances for each sample are the same D. Both samples were random and independent from one another
B. both samples were random and are dependent on one another
What deals with both variants between and within groups? A. Related groups t-test B. Independent groups t-test C. Correlated groups t-test D. All of the above
B. independent groups t-test
Why do scientists use = .05 as the error probability? A. Karl Pearson made it up B. It was 100% arbitrary and decided while they were drunk and smoking cigars C. Out of every 100 experiments that took place during the 1800s, there were 5 mistakes
B. it was 100% arbitrary and decided while they were drunk and smoking cigars
Most distributions are ___________. A. Normally distributed B. Not normally distributed
B. not normally distributed
Which of these is not a type of sampling distribution? A. Population distribution B. Size distribution C. Sampling distribution D. Sample distribution
B. size distribution
Standard error of mean tends to be: A. Larger than standard deviation B. Smaller than standard deviation C. About the same D. They are not comparable
B. smaller than standard deviation
What happens to the distribution the more you sample? A. The more you sample, the more the distribution becomes leptokurtic B. The more you sample, the more the distribution becomes normal C. The more you sample, the more the distribution becomes mesokurtic D. The distribution does NOT change with the size of sampling
B. the more you sample, the more the distribution becomes normal
Why can't we use one sample to estimate a population? A. Most samples are not random B. They are not an accurate representation of the population C. Samples are biased D. The population is not an accurate representation of the sample
B. they are not an accurate representation of the population
What are Z-scores used for? A. To describe the relationship strength between two variables. B. To figure out how a sample compares to the population. C. To estimate the difference between variance of individuals. D. Whether two groups are correlated
B. to figure out how a sample compares to the population
When do you use a t-test? A. When you don't have a population, but have two large sample groups B. When you don't have a population, but have two small sample groups C. When you have a population and one sample group D. When you compare two population means
B. when you don't have a population, but have two small sample groups
What do we use to figure out how a sample compares to the population? A. t-test B. Z-score C. Null hypothesis D. p-value
B. z-score
In a standard normal distribution, the area underneath the curve is equal to: A. Greater than 1.0 B. Less than 1.0 C. 1.0 D. Equal to 10
C. 1.0
Which of these statements is NOT true about standard error? A. Indicator of how reliable the mean is B. It is the standard deviation of the population mean C. A larger standard error indicates that a given mean is an accurate reflection of the population
C. a larger standard error indicates that a given mean is an accurate reflection of the population
How is a sample distribution and a population distribution different? A. they are the same thing B. a sample distribution shows scores for a population and a population distribution shows scores for a sample C. a sample distribution shows scores for a sample and a population distribution shows scores for a population D. a sample distribution shows a distribution of sample statistics and a population distribution shows scores for a population
C. a sample distribution shows scores for a sample and a population distribution shows scores for a population
How can you increase your statistical power? A. Increase sample size B. Reduce error variance C.Both a and b D. None of the above
C. both a and b
What's not true about the difference between the t-test for independent samples, and t-test for correlated samples? A. The independent t-test has less power than the correlated one. B. The independent t-test has more variance than the correlated one. C. Both are correct. D. Both are incorrect
C. both are correct
How are t-tests different from z-tests? A. T-tests do not use a sample mean, while z-tests do B. T-tests don't determine significance while z-tests do C. T-tests do not use a population standard deviation, while z-tests do D. T-tests use a population standard deviation, while z-tests do not
C. t-tests do not use a population standard deviation, while z-tests do
According to the central limit theorem, what happens to a sampling distribution as n gets larger? A. the distribution gets wider B. the distribution becomes negative C. the distribution becomes more normal D. the distribution requires a q-test in order to be analyzed
C. the distribution becomes more normal
Which of these assumptions do you need to make when conducting t-test? A. The variances for each sample differ from each other B. The samples are dependent on each other C. The mean difference is sampled from a normal distribution of mean differences D. The samples come from the same population
C. the mean difference is sampled from a normal distribution of mean differences
Which of the following is the best example of a good sample: A. A research study wants to know how do people 65+ move around town, so they sample a neighborhood and recruit university students B. A researcher asked members at a family reunion to fill out a survey about availability of dog parks in the community C. A researcher does not inform participants that the treatment could harm them D. A researcher recruits students from top universities to gain insight into what stress looks like for them
D. a researcher recruits students from top universities to gain insight into what stress looks like for them
Standard error... A. Is an indicator of how reliable the mean is B. Is the standard deviation of the sampling distribution C. Is used with a sampling distribution D. All of the above
D. all of the above
What is an example of a correlated samples t-test? A. Pretest/Posttest B. Couples C. Related samples D. All of the above
D. all of the above
When would you use a correlated-samples T-test? A. Pre-test/post-test B. Related samples C.Couples D. All of the above
D. all of the above
Which of the following is true about the standard error of the mean? A. It is typically smaller than the population standard deviation B. It is typically smaller than the sample standard deviation C. It is the standard deviation of the sampling distribution D. All of the above
D. all of the above
Which of the following is not one of the caveats of using a t-test for independent samples? A. Assume the mean difference comes from a normal distribution. B. Assume homogeneity of variance. C. Assume both samples are random and independent. D. Assume the sample size is larger than 100.
D. assume the sample size is larger than 100
According to the central limit theorem, as N increases, the distribution: A. Skews to the right B. Skews to the left C. Becomes uniform D. Becomes more normal
D. becomes more normal
When is it a good idea to use the the t-test? A. When we do not have the population standard deviation B. When we have a large sample size C. When we are comparing smaller samples to each other D. Both a and c
D. both a and c
What is true about the population parameters? A. It is used in a t-test. B. It is used in a z-test. C. Both are correct. D. Both are incorrect.
D. both are incorrect
In relation to normal distributions, you can use Z scores for all of the following EXCEPT: A. Determine where a sample stands in relation to the population B. Calculate percentiles C. Calculate probabilities D. Determine causality
D. determine causality
What type of samples in two-sample tests have more statistical power? A. Different samples B. Independent samples C. Homoscedasticity samples D. Matched/paired/correlated samples
D. matched/ paired/ correlated samples
Thanks to the Central Limit Theorem, we know when N becomes larger, the sampling distribution becomes: A. Leptokurtic curve B. Mesokurtic curve C. Platykurtic curve D. Normal distribution
D. normal distribution
What's the difference between standard error and standard deviation? A. Standard deviation is an indication of how reliable the mean is B. A smaller standard deviation indicates that a mean is an accurate reflection of the population C. A larger standard deviation indicates that a mean is an accurate reflection of the population D. Standard deviation is an indicator of how far data differs from the mean on average
D. standard deviation is an indicator of how far data differs from the mean on average
What does the line of best fit represent? Line that minimizes the difference between the predicted Y and actual Y observed Line that maximizes the difference between the predicted Y and actual Y observed The significance of the difference between two variables
Line that minimizes the difference between the predicted Y and actual Y observed
What is the correct definition of statistical power? There is a strong difference due to chance The probability of rejecting the null hypothesis where it is in fact false There is a weak difference due to chance The probability of rejecting the null hypothesis where it is in fact true
The probability of rejecting the null hypothesis where it is in fact false
When would you use a two-tail test? To test if there is a significant difference between your sample and the population To test if there is an insignificant difference between your sample and the population To test if there is a significant increase between your sample and the population To test if there is an insignificant increase between your sample and the population
To test if there is a significant difference between your sample and the population
In a study, a scientist falsely claims that coffee leads to cancer. What kind of error is this? Type I Type II Type III Not enough information
Type I
You're a scientists trying to find out whether your class has higher average test grades than all other classes at UNC. Which test could you use? One-tailed Two-tailed Neither Either
either
Why can correlation not imply causation? Correlation does imply causation Correlation shows the strength of the relationship between two variables The regression equation isn't standardized Extraneous variables can result in correlational relationships
extraneous variables can result in correlational relationships
The null hypothesis states that the difference are NOT due to chance. True False
false
True or False: In general, scientists are more worried about Type II errors than Type I errors. True False Not enough information
false
Which of the following is NOT an assumption of regression or correlation? Interval or ratio scales Linear relationship Normal distributions for X and Y Heteroscedasticity
heteroscedasticity
What is the definition of a Type 2 Error? Correctly rejects the null hypothesis Incorrectly rejects the null hypothesis Correctly accepts the null hypothesis Incorrectly accepts the null hypothesis
incorrectly accepts the null hypothesis
What is the definition of a Type 1 Error? Correctly rejects the null hypothesis Incorrectly rejects the null hypothesis Correctly accepts the null hypothesis Incorrectly accepts the null hypothesis
incorrectly rejects the null hypothesis
Which of the following is NOT true of regression? It doesn't tell us the relative strength of a relation It doesn't allow us to truly compare relations between variables It tells us the relative strength of a relation It shows the best line for describing the relationship between two variables
it tells us the relative strength of a relation
Regression allows us to predict outcomes assuming relationship is _____. Parabolic Curvilinear Linear Asymptomatic
linear
Which of the following is NOT true for the line of best fit? Minimizes differences between Y and the predicted Y Maximizes differences between Y and the predicted Y Shows correlation between two variables Used to identify trends occurring within the dataset
maximizes differences between Y and the predicted Y
The critical Z-score for a two-tailed test with α=0.05 is... 1.96 -1.96 1.65 Must be between 1.96 and -1.96
must be between 1.96 and -1.96
If we wanted to see if the average GPA at UNC is better than Duke, what type of test would we use? One-tailed Two-tailed Three-tailed None of the above
one-tailed
Which of these choices best describes the definition of R-squared? Slope of the regression line Percent of Y predicted by X The correlation coefficient Percent of X predicted by Y
percent of Y predicted by X
Given the following regression equation, what does Y-hat represent? Ŷi = bXi + a Y-intercept Observed Y value Predicted Y value Slope
predicted Y value
Regression must have at least two variables. What are these variables? Predictor and moderator Moderate and outcome Predictor and outcome Predictor moderator and outcome
predictor and outcome
What happens to our regression equation (slope) when we standardize? Slope becomes correlation (r) Slope becomes r2 Slope remains the same
slope becomes correlation (r)
What value should you use to represent the slope of a regression line when dealing with multiple regression? B (unstandardized) Standardized Coefficient Beta t
standardized coefficient beta
What is a residual? The difference between the actual and predicted point The sum of the actual and predicted point The difference between the independent and dependent variable The sum of the independent and dependent variable
the difference between the actual and predicted point
What do residual scores represent? The predicted value subtracted from the observed value The observed value subtracted from the predicted The addition of all the variables together Both A and C
the predicted value subtracted from the observed value
Select the true statement. Type II Error occurs when a true null hypothesis is rejected The alpha level changes based on the sample size There are only two types of error in hypothesis testing Hypothesis testing is unethical
there are only two types of error in hypothesis testing
True or False: As your sample gets bigger your t-score value gets close to a z-score. True False
true
If you reject the null hypothesis that means that... You can say there is significance You can say there is no significance You can say the results are not likely due to chance You cannot say the results are due to chance
you can say the results are not likely due to chance
How can you change an alpha level? You can't change an alpha level Increase sample size Use a two-tailed test Reduce error variance
you can't change an alpha level