stats exam 3
In factorial notation, the expression 6! means:
(6)(5)(4)(3)(2)(1) = 720
If you draw a line of best fit through a scatterplot, and your line slopes sharply upward from left to right, like this: / Which correlation coefficient most closely represents that relationship?
+.85
The standard error of the mean:
- decreases as sample size (N) increases, all other things being equal. - is the standard deviation of the sampling distribution of the mean. - is the "average" amount of sampling error across all random samples of a given size.
Correlations for the types of variables studied by behavioral scientists will seldom exceed _____ and will often be considerably smaller.
-.40 or +.40
Which correlation coefficient most closely represents the relationship depicted in the scatterplot below?
-.75
Julie took the same psychology exam as John and received a score of 65. What is her standard score?
-2.5
The average family watches 40 hours of TV a week and the SD is 4. Jane says she watches about 28 hours of TV a week and her variance is 4. What is her standard score?
-3
Which of the following correlation scores represents the weakest relationship?
.09
What would the value of eta-squared be if an independent t-test yielded an observed t value of -1.73 and the number of subjects in the placebo group was 10 and the number of subjects in the treatment group was 10?
.142
If the scores of a class at midterm are 70, 70, 70, 80, 80, 90, 92, and 92, what is the relative frequency of the score 80?
.25
A sample of 500 people aged 30 - 40 years old were interviewed to study the relationship between whether a person graduated from college or only high school and whether s/he changed jobs in the last 5 years. The results are found in the contingency table below. Answers below have been rounded to two decimal places. What is the probability that an individual has changed jobs?
.49
In a population consisting of 60 males and 40 females, the probability of randomly selecting a male is:
.60
A sample of 500 people aged 30 - 40 years old were interviewed to study the relationship between whether a person graduated from college or only high school and whether s/he changed jobs in the last 5 years. The results are found in the contingency table below. Answers below have been rounded to two decimal places. What is the probability that an individual has graduated from college?
.64
As a rough guide, investigators generally attempt to achieve statistical power (the probability of correctly rejecting the null hypothesis when it is false) in the range of _____ to _____.
.80 to .95
What proportion of scores in a normal distribution lies between z scores of -1.55 and +1.55?
.8788
Dr. August is investigating whether or not women have more job satisfaction after they are given praise or a gift card (different levels of rewards). She has recruited 10 women from Intel and for 6 months the women received praise then the next 6 months they received a gift card. The conditions were counterbalanced and a survey on job satisfaction was distributed following the treatment conditions or every 6 months until the research was completed. After finishing her study Dr. August was busy writing a grant, therefore she has asked you to analyze her data below. Please note the survey was based on 1-5 Likert scale with high scores indicating high job satisfaction and low score indicating dissatisfaction. What is the strength of the relationship of the data above?
.90
A slope of _____ would be represented by a _____ line.
0; horizontal
Given a set of outcomes that are mutually exclusive and exhaustive, the sum of the probabilities of the outcomes will always equal:
1.0
For an alpha level of .05, an unexpected result in the one sample z test includes any sample mean occurring more than _____ standard errors below or more than _____ standard errors above the value of mu (the population mean) represented in the null hypothesis.
1.96
A biopsychologist is interested in how quickly rats learn a complicated maze. He has 5 rats. Their names are Alfred, Bob, Cliff, Dave, and Eli. He wants to run 2 rats through a maze at a time. How many subsets of 2 rats could he choose to run through the maze if order does NOT matter?
10
If you have 5 questionnaires you want to hand out to a group of people, how many combinations of these 5 questionnaires, taking 3 at a time, are there? (Order in which the questionnaires are given to the person does not matter.)
10
A researcher wanted to compare eye color distributions of babies born with a birth defect with babies who did not have the birth defect. 26 babies with the defect were matched on gender, race, and age with 52 babies without the defect. The frequency of different eye colors as a function of the presence or absence of the defects was as follows: The marginal frequencies for eye color are
16, 19, 43
Dr. Hind used a One-Way ANOVA, between subjects design and has set his alpha level at .05 nondirectional. He reported the following: F(2, 19)=6.99. In Dr. Hind's study, the df associated within groups is ________ and the df associated between groups is ________.
19, 2
An independent groups t-test has ________ of the independent variable.
2 levels
An independent groups t-test has _____levels of the independent variable.
2 levels
A sample of 500 people aged 30 - 40 years old were interviewed to study the relationship between whether a person graduated from college or only high school and whether s/he changed jobs in the last 5 years. The results are found in the contingency table below. Answers below have been rounded to two decimal places. Which of the following is a marginal frequency in the above table?
245
A physical therapist is interested in whether people with different bone structures will be affected differently by her new therapy program. She randomly chooses two groups of patients based on bone type: small-boned frame (XS) and large-boned frame (XL). Patients all had similar injuries. She measures the number of days each person takes to progress to a predetermined point. Here is the data: Small Bones Large Bones Mean 10.00 12.00 n 15 12 s2 4.5 5.0 SS 108 105 What are the degrees of freedom for the t test for two independent samples?
25
Dr. Hinson wants to investigate if rats fed a diet enriched with omega will do better at a maze learning task then when these same rats are given a normal diet. He has a total of 30 rats for his experiment. In Dr. Hinson's study if he is using a total of 30 rats what is his degrees of freedom?
29
Dr. Qui found his test scores to be as follows 60, 92, 75, 79, 75, and 90. What is the range of the test scores?
32
What proportion of scores are equal to a SD=1 in a normal distribution?
34.13%
Dr. Johnson is studying the effects of aspirin on heart rate. He believes that aspirin will lower heart rate. He has 2 groups, the treatment group receives 80mg of aspirin for a week (n=4), the control group received a placebo (n=4), and their heart rates are compared. Below is his data in which heart rate is beats per minute. He sets his alpha level at 0.05 and uses a directional test. What is the t-value for the data above?
4.20
If the sample sizes in the two groups are each greater than _____ and roughly comparable, then the independent groups t test is robust to rather severe departures of the _____ assumption.
40; normality
Dr. Johnson is studying the effects of aspirin on heart rate. He believes that aspirin will lower heart rate. He has 2 groups, the treatment group receives 80mg of aspirin for a week (n=4), the control group received a placebo (n=4), and their heart rates are compared. Below is his data in which heart rate is beats per minute. He sets his alpha level at 0.05 and uses a directional test. What is the 95% confidence intervals for Dr. Johnson's study?
5.78 to 15.72
An experiment is conducted with 200 trials and the probability of a success in each trial is .250. What is the mean of the binomial distribution?
50.00
An instructor has given his final exam and scores that are normally distributed. Nancy scores a 90 and Ronald scores a 75. The class mean on the exam was 85 and the standard deviation was 36. What is Nancy's T score? (Hint use formula: mean=50 +10 (standard or Z score)
51.4
A teacher wants to determine the mean score of students taking a newly developed test. Her students scored as follows: 40, 45, 60, 65, 70, 70, 70, 80, 90, 100. Compute the mean for the class.
69
If you were provided a set of 8 scores, ____ would correspond to the degrees of freedom associated with the sum of squares of the 8 scores
7
A deck of cards has 52 cards. Each deck has 4 aces, 4 kings, and 4 queens. What is the probability of drawing an ace from a full deck of cards? (The rules for rounding apply.)
8% chance of getting an ace
In an APA results section and independent groups t-test is written as follows t(82) = 5.66, p<0.05. Based on this information what are the degrees of freedom?
82
Outliers
A case or set of cases that shows a very extreme score relative to the majority of cases in the data set—so extreme that the score is suspect.
Alternative Hypothesis
A competing proposal to the null hypothesis. If the null hypothesis states that the number of events one should observe is 10, then the alternative hypothesis might state that the number of events one should observe is not 10.
Monte Carlo Study
A computer simulation study in which the statistician uses the computer to generate scores for hypothetical populations that he or she knows violates a distributional assumption in some way. Then the test statistic in question is applied many times to random samples from the population to determine the affects of assumption violations on the frequency of Type I and Type II errors.
Correction for Continuity
A correction factor for applications of the binomial distribution for cases of small number of events to create a closer approximation to the normal distribution.
Sampling Distribution of the F Ratio
A distribution of F ratios calculated on all possible random samples of a given size based on the levels of the independent variable.
Sampling Distribution of the Mean
A distribution of mean scores, consisting of the means for all possible random samples of a given size from a population.
Probability Distribution
A distribution of probabilities across values of a variable
probablity distribution
A distribution of probabilities across values of a variable.
Normal Distributions
A family of theoretical distributions that are defined by a mathematical formula. There is a different normal distribution for every combination of mean and variance. All normal distributions are symmetrical and bell shaped. The mean is equal to the median which is equal to the mode in a normal distribution.
Bar Graph
A frequency histogram for qualitative variables; bars are drawn such that they do not touch one another.
Scatterplot
A graph for visualizing the relationship between two quantitative variables. Scores for one of the variables (X) are listed on the abscissa and scores for the other variable (Y) are listed on the ordinate. A dot is placed in the graph where the X and Y scores intersect.
frequency histogram
A graph of frequency distribution for a quantitative variable in which frequencies occur on the ordinate and scores occur on the abscissa. The frequencies are denoted with bars that occur side by side.
Control Group
A group of participants in an experiment that are not exposed to any experimental treatment.
Linear Model
A mathematical formula that identifies parameters that define a linear relationship = a + b X.
Scales
A measuring device or instrument used to obtain a set of measures.
probability density function
A method for representing the probability of observing scores between any two values in the context of a probability distribution
In order to determine the nature of the relationship with a one-way ANOVA: _____.
A multiple comparison procedure or post-hoc test must be used
Tukey HSD Test
A multiple comparison procedure used to compare all possible pairwise differences between population means.
Mutually Exclusive
A mutually exclusive category is one which a given individual can be classified in one and only one category. For example, if the category is gender, then a person can be classified as male or female, but not both.
Slope
A parameter in the linear model. It indicates the number of units that the Y variable is predicted to change given a one unit change in X.
Intercept
A parameter in the linear model. It is the predicted value of Y when X=0.
Variables
A phenomenon that takes on different values, or levels.
Statistically Nonsignificant
A phrase that means that the null hypothesis was not rejected based on the evaluation of a test statistic.
Statistically Significant
A phrase that means that the null hypothesis was rejected based on the evaluation of a test statistic.
Alpha Level
A probability value that is used to determine whether a result can be attributed to chance or non-chance in the context of hypothesis testing.
Alpha Level
A probability value that is used to determine whether a result can be attributed to chance or non-chance in the context of hypothesis testing. The alpha level also reflects the probability of making a Type I error.
Confidence Interval
A range of values within which we are relatively confident that the population parameter occurs. The confidence interval is stated in terms of a percentage (e.g., 95%) and indicates the percentage of times that the population mean falls within the confidence interval as computed in the sample data for all possible random samples of a given size.
Variance Ratio
A ratio of two variance estimates.
precentage
A relative frequency multiplied by 100.
Biased Estimator
A sample statistic whose average (mean) over all possible random samples of a given size does not equal the value of the corresponding population parameter. If the average is larger than the value of the population parameter, the estimator is said to be positively biased. If the average is smaller than the value of the population parameter, the estimator is said to be negatively biased.
Unbiased Estimator
A sample statistic whose average (mean) over all possible random samples of a given size equals the value of the corresponding population parameter.
Representative Sample
A sample that has the same basic characteristics as the population on a set of variables of interest.
Nullified Score
A score on the dependent variable in which the effect of the independent variable has been removed.
Standard Scores
A score that indicates the number of standard deviations that a raw score is above or below the mean.
Correlation Matrix
A set of correlation coefficients organized into rows and columns which shows correlations between all possible combinations of variables.
Combination
A set of objects or events in which the ordering of objects is irrelevant. AB is the same combination as BA.
Exhaustive
A set of outcomes that includes all the possible outcomes that could occur.
Sampling Distribution of the Difference Between Two Independent Means
A set of scores that are mean differences between two independent samples for all possible pairs of samples of given sizes.
Summation Notation
A shorthand way of writing an instruction to sum a set of scores or perform other algebraic manipulations on the scores.
z Score
A standard score in a normal distribution
T Score
A standard score that has been transformed so that it has a mean equal to an a priori specified value and a standard deviation equal to an a priori specified value. A T score avoids confusion associated with decimals and negative values.
Hypothesis
A statement proposing that something is true about a given phenomenon.
Sample Statistic
A statistical index based on data from a sample that can be used to infer something about a population or that can be used to describe a sample.
Test Statistic
A statistical index calculated from sample data that is used to formally conduct a null hypothesis test (i.e., to evaluate the reasonableness of a sampling error interpretation of sample discrepancies from the null hypothesis). In this chapter, the z score and the t score are examples of test statistics.
Population Parameter
A statistical index that describes a population based on all data from that population.
Pearson Correlation Coefficient
A statistical index that reflects the degree to which two variables approximate a linear relationship. Values can range from -1.00 to + 1.00. The larger the absolute value of Pearson correlation, the closer the approximation to a linear relationship.
Eta-Squared
A statistical index that reflects the proportion of variability in the dependent variable that is associated with the independent variable for the sample data.
Levene Test
A statistical procedure for testing a null hypothesis that two population variances are equal.
One-Sample t Test
A statistical procedure that involves the calculation of a test statistic (a t score) that has a sampling distribution that closely approximates a t distribution. The test statistic is used in conjunction with our knowledge of the t distribution to reject or not reject the null hypothesis.
Independent Groups t Test
A statistical procedure used to test a null hypothesis of equal population means when the independent variable is between-subjects in nature and has two values and the dependent variable is quantitative.
Correlated Groups t Test
A statistical procedure used to test a null hypothesis of equal population means when the independent variable is within-subjects in nature and has two values and the dependent variable is quantitative.
Regression
A statistical technique that can be used to identify a "best fitting" line that describes the linear relationship between two quantitative variables
One-Way Between-Subjects Analysis of Variance
A statistical technique that is used to analyze the relationship between two variables, under the same circumstances as the independent groups t test except that the independent variable has more than two levels. The procedure is focused on testing the null hypothesis of equivalent population means across the levels of the independent variable.
One Sample z Test
A statistical test used to test the viability of the null hypothesis that the population mean is equal to a specific value. It is applied in cases where the population variance is known, which is rarely the case.
Random Sampling
A strategy for selecting samples from a population. Every member of the population has an equal chance of being selected for the sample.
Matching
A strategy that is used to control for the adverse effects of confounding variables and disturbance variables. In this approach, an individual in one group is "matched" with an individual in each of the other groups such that all of these individuals have the same value on the confounding/disturbance variable.
Holding a Variable Constant
A strategy used to control for the effects of a disturbance variable or a confounding variable. It involves selecting research participants in such a way that they all have the same value on the variable being held constant (e.g., for gender, only including females in the study).
Matched-Subjects Designs
A strategy where individuals at different levels of the independent variable are matched and then the data are treated as if it is from a within-subjects design.
Cross Sectional Design
A study where age is a between-subjects independent variable.
Longitudinal Design
A study where age is a within-subjects independent variable.
Sample
A subset of the population.
Sum of Squares Between
A sum of squares as applied to a set of scores for each individual in the experiment, where a given individual's "score" is the mean value of the group that the individual is a member of. It reflects between-group variability.
Sum of Squares Within
A sum of squares as applied to deviation scores from the group mean across all individuals in the study. It reflects within-group variability.
Sum of Squares Total
A sum of squares as applied to the dependent variable across all individuals in the study.
Sum of Squares Error
A sum of squares calculated on scores on the dependent variable after the effects of the independent variable have been statistically removed (i.e., on nullified scores).
Sum of Squares Total
A sum of squares calculated on the dependent variable across all individuals in the experiment.
Sum of Squares Explained
A sum of squares calculated the index of the explained variance.
normal distribution
A symmetrical distribution that is characterized by a "bell shape."
grouped frequency distribution
A table of a frequency distribution, however scores on the variable of interest are grouped together and "treated as one."
Contingency Table
A table of numbers in which the columns are the values of one variable and the rows are the values of a second variable. The entry in a given cell of the table is the frequency with which the combination of values defined by the column and row occurs.
Random Number Table
A table of numbers that have been generated by a computer so that they closely approximate a set of random numbers.
frequency distribution
A table that lists scores on a variable in order and shows the number of individuals who obtained each score or value.
Summary Table
A table that summarizes all of the calculations for applying the F test. The table includes the sum of squares between, the sum of squares within, the sum of squares total, the degrees of freedom between, the degrees of freedom within, the degrees of freedom total, the mean square between, the mean square within, and the F ratio.
Robust
A term that indicates how sensitive a statistical test is to violations of its underlying assumptions. A statistical test that is "robust" is said to not be affected (in terms of the incidence of Type I and Type II errors) by violations of its underlying assumptions.
Central Limit Theorem
A theorem that describes characteristics of a sampling distribution of the mean Given a population with a mean of μ and a standard of σ, then the sampling distribution has a mean of μ and a standard deviation of σ / sqrt(?N) [where sqrt means "take the square root of "] and approaches a normal distribution, as the sample size approaches infinity.
Sampling Distribution of the Mean of Difference Scores
A theoretical distribution consisting of mean difference scores across individuals for all possible random samples of a given size that could be selected from a population.
F Distribution
A theoretical distribution that has a known form and for which probability statements can be made about the occurrence of different ranges of scores.
t Distribution
A theoretical distribution that is bell shaped and symmetrical, much like the normal distribution. Statisticians are able to make probability statements about the occurrence of different ranges of values in a t distribution, making it useful for statistical inference.
stem and leaf plot
A type of frequency graph in which scores are listed on the ordinate and represented in such a way that every score in the distribution appears on the graph.
Confounding Variable
A variable that is related to the independent variable (the presumed cause) and that affects the dependent variable (the presumed effect), rendering a causal inference between the independent variable and the dependent variable ambiguous.
Disturbance Variable
A variable that is unrelated to the independent variable (and hence, not confounded with it), but that affects the dependent variable.
F Ratio
A variance ratio named for the statistician Sir Ronald Fisher. In the present chapter it is the ratio of the mean square between divided by the mean square within.
Boxplot
A way of presenting measures of central tendency and variability graphically. It presents a box, the height of which is related to the amount of variability.
If you were studying the effects of age differences on reaction times, comparing teenagers to middle aged people to elderly people, which axis should have the age groups on it, and which axis should have reaction time?
Age would go on the X axis, and reaction time on Y.
Statistics relevant to the analysis of Bivariate relationships will focus on questions including ________
All of these are correct. // both inference about the relationship and strength of the relationship strength of the relationship nature of relationship inference about the relationship
In order to select the correct statistical test you need to know if the experimental design is:
All of these are correct. // scale of measurement of the dependent variable. the number or levels of the independent variable. between or within subjects design. both between or within subjects design, and scale of measurement of the dependent variable.
A correlated -groups t test can be used to analyze the relationship between two variables when _____.
All of these// the dependent variable is measured on a ratio level the independent variable is within-subjects the independent variable has only two levels
Central Tendency
An average of a set of scores; a value around which other scores cluster.
Carry-Over Effects
An effect that can occur in within-subjects designs. It occurs when the research participant's experience in the first condition s/he participates in affects performance in the second condition.
_____ strategy is one where a set of procedures or manipulations is performed in order to create different values of the independent variable for the research participants.
An experimental
Standard Error of Estimate
An index of predictive error that represents an average discrepancy between the predicted and observed Y scores.
Permutation
An ordered sequence of a set of objects or events. AB and BA are two different permutations or orderings.
Standard Deviation Estimate
An unbiased estimator of the population standard deviation.
Variance Estimate
An unbiased estimator of the population variance.
Box and Whisker Plot
Another name for boxplot.
Independent Groups Designs
Another word for between-subjects designs.
Correlated Groups Designs
Another word for within-subjects designs.
Repeated Measures Designs
Another word for within-subjects designs.
Mean Square
Any sum of squares divided by its associated degrees of freedom.
Random Assignment
Assigning people to groups using random procedures so that a person is equally likely to be assigned to any of the experimental or control conditions.
Parametric statistics:
Assume that scores of the variable being measured are normally distributed in the population
Which of the following is NOT one of the steps of hypothesis testing used when comparing two independent means?
Assuming the alternative is true, characterize the mean and standard deviation of the sampling distribution of the differences between two independent means
Dr. Nan is studying the effects of Ginseng on recall. He has 2 groups of subjects (20 subjects in each group). Subjects are given Ginseng or a Placebo and then asked to recall a list of words they previously studied. However, to avoid confounds what should Dr. Nan do?
Both match groups on an IQ test and assign to groups, and randomly assign subjects to groups.
In writing up the statistical results of a study, which statement is most likely to appear in a research report?
Both the hypothesis was confirmed at the .01 level, F(2)=9.90, and the strength of the relationship, as indicated by eta-squared, was .45.
Nonparametric Statistics
Class of statistics that tend to focus on medians, interquartile ranges, and ordinal level measures and make minimal, if any, assumptions about the distribution of scores in a population.
Pooled Variance Estimate
Combining two samples to increase the degrees of freedom on which the estimate is based and thereby obtain a better estimate.
Dr. Hinson wants to investigate if rats fed a diet enriched with omega will do better at a maze learning task than when these same rats are given a normal diet. He has a total of 30 rats for his experiment. What should Dr. Hinson do before he starts his experiment?
Counterbalance the diet so 15 rats receive omega then normal chow and 15 normal chow then omega diet.
Partitioning of Variance
Decomposing the total variability into different components or sources. The sum of squares total can be broken down into two parts, the sum of squares between and the sum of squares within.
theoretical distributions
Distributions of scores based on a mathematical function.
Empirical Distributions
Distributions of scores based on actual measurements collected in the real world.
Observational Strategy
Does not involve the process of actively creating values on an independent variable, but rather involves measuring differences in values that naturally exist in the research participants.
Cell
Each unique combination of values of two variables in a contingency table.
Event
Each unique outcome from a trial.
Intelligence test scores are considered to be aNo ____ scale of measurement.
Either ordinal or interval; there is support for both
Independence
Event A is independent of event B if the occurrence of B does not affect the probability of the event A.
Based on the Method of Presentation section of Chapter 12, suppose a researcher is reporting the results of a one-way ANOVA. The independent variable has 4 levels, and there are 30 subjects in each group. If he obtained an F value of 2.91, using an alpha level of .05, how would he report this in a psychological journal?
F (3, 26) = 2.91, p < .05
Type II Error
Failing to reject the null hypothesis when it is false.
cumulative frequencies
Frequency associated with a given score plus the sum of all frequencies for scores less than that score.
If a researcher is interested in having a group of 100 people complete a survey and she plans to calculate a variance estimate and standard deviation estimate for that group, how can she increase her degrees of freedom before she does her calculations?
Have more than 100 people complete the survey.
Dependent Variable
In a causal relationship between two variables, the dependent variable is the presumed effect of the other variable. It is the variable that is "being influenced."
Independent Variable
In a causal relationship between two variables, the independent variable is the presumed cause of the other variable. It is the variable that is "doing the influencing."
Binomial Expression
In a sequence of n independent trials, each of which has only two possible outcomes (arbitrarily called a "success" and a "failure"), with the probability of p of success and the probability q of failure, the probability of r successes in n trials is defined by the binomial expression (see equation 6.12 in the text).
Which of the following statements is true of a median score?
It divides a distribution in half.
Police officers must take an exam before they are hired in the department. Jack's has a T-score on the exam was 50 and John's T-score was 60. Who will most likely get hired?
Jack
What statement below is TRUE in regards to a normal distribution?
Mean, median and mode are always equal
The characteristic bell shape of the normal distribution occurs because:
Most people score in the average range, in the center of the distribution, AND fewer people score in the less average ranges, at the two tail-ends of the distribution.
A sum of squares around a sample mean will always have _____ degrees of freedom associated with it.
N - 1
In the correlated groups t test, the relevant t distribution has _____ degrees of freedom associated with it.
N -1
Suppose a researcher is reporting the results of an independent groups t test. If he obtained a statistically significant t value of 2.5 with 15 subjects in each group using an alpha level of .05, how would he report this in a psychological journal?
NOT t (14) = 2.5, p < .05 // not t (14) = 2.5, p > .05
_____ statistics focus on differences between distributions of scores and can be used to analyze quantitative variables that are measured on an _____level.
Nonparametric; ordinal
Statistics
Numerical indices based on data from a sample.
Parameters
Numerical indices based on data from an entire population.
The 95% confidence interval for the mean differences is reported following a(n) ________.
Pearson correlation
Real Limits
Points falling one-half a measure unit above a number (the upper real limit) and one-half a measure unit below that number (the lower real limit).
Categorizing cats by the color of their fur (white, orange, black, brown, and multicolor) is an example of:
Qualitative variables using a nominal scale
Sampling Error
Random or systematic factors that yield a sample statistic that does not equal the value of its corresponding population parameter.
Cohort Effects
Refers to historical variables that are confounded with age and therefore serve as alternative explanations to developmental processes.
Type I Error
Rejecting the null hypothesis when it is true.
Positive Relationship
Relationship in which as scores on X increase, scores on Y also increase.
Negative Relationship
Relationship in which as scores on X increase, scores on Y decrease.
Sampling With Replacement
Replacing cases back into the population before another case is randomly selected.
Between-Subjects Designs
Research designs involving between-subjects independent variables. A between-subjects independent variable is one where a research participant who has a given value of the independent variable cannot also have a different value on the independent variable. Stated more informally, different individuals are used for each value or level of the independent variable. For example, if the independent variable is gender, then the individuals who have the value "female" are not the same individuals who have the value "male."
Within-Subjects Designs
Research designs involving within-subjects independent variables. A within-subjects independent variable is one where the same individual is used at each level or value of the independent variable. For example, if a study is examining the effects of alcohol on learning in a recognition task, the independent variable might be the presence or absence of alcohol. An individual would first perform the recognition task-learning test while not under the influence of alcohol and then again, while under the influence of alcohol.
One-Tailed Test
Same as a directional test.
Inverse Relationship
Same as a negative relationship.
Two-Tailed Tests
Same as a nondirectional test.
Direct Relationship
Same as a positive relationship.
Error Variance
Same as unexplained variance.
In a one-way analysis of variance, what factors influence within group variability? (Read all answers and then choose the best one.)
Sampling error AND individual differences
In a one-way analysis of variance, what factors influence between group variability? (Read all the answers before choosing the best one.)
Sampling error AND treatment effects
Sampling Without Replacement
Selecting a case at random and then, without replacing this case, selecting another case at random.
frequency polygon
Similar to a frequency histogram, however bars are not used, but rather, solid dots corresponding to the appropriate frequencies are placed directly above the score values. The dots are connected with a solid line and the polygon is "closed" on the abscissa with boundary frequencies of zero.
line plot
Similar to a frequency polygon, except the line connecting the dots is not "closed" on the abscissa. Rather, the left and right most portions of the line end on the lowest and highest scores, respectively.
Constant
Something that does not vary within given constraints.
Which of the following is NOT one of the three conceptualizations of probability discussed in the text?
Standard scores derived from probabilities
Nondirectional Test
Statistical tests that are designed to detect differences from a hypothesized population value (as stated in the null hypothesis) in both directions (i.e., the population mean is greater than the value specified in the null hypothesis or it is less than the value specified in the null hypothesis).
Directional Test
Statistical tests that are designed to detect differences from a hypothesized population value (as stated in the null hypothesis) in only one direction (e.g., the population mean is greater than the value specified in the null hypothesis).
Multiple Comparison Procedures
Statistical tests used to determine the nature of the relationship after the omnibus null hypothesis has been rejected for the overall F test.
If a score is very extreme relative to the majority of other scores in the dataset, this is important for the researcher to know because:
Such scores are suspect and could be the result of an error.
The greek letter "sigma" is used in statistics to mean:
Summation
Parametric Statistics
Techniques that involve the analysis of means, variances, and standard deviations and require quantitative dependent variables. The methods are usually applied when these variables are measured at a level that at least approximates interval characteristics. They make assumptions about the distributions of scores in a population.
Unsigned Deviation Scores
The absolute value of the deviation score.
Trial
The act of creating the event.
Hypothesis Testing
The act of specifying an expected result of a study assuming the null hypothesis is true. If our observations are sufficiently discrepant from the expected result that the difference cannot be attributed to chance, then the null hypothesis is rejected. Otherwise, we fail to reject the null hypothesis.
Population
The aggregate of all cases to which one wishes to generalize statements.
Density Curve
The area under the curve in a probability distribution representation of a continuous variable.The area under the curve in a probability distribution representation of a continuous variable.
Normality Assumption
The assumption that scores on the variable are normally distributed in the population.
Homogeneity of Variance
The assumption that the population variances are equal.
Least Squares Criterion
The criterion used to determine the values of the slope and intercept when trying to fit a linear model to a set of data. It minimizes the sum of the squared discrepancies between the predicted and observed Y scores.
Critical Values
The critical values are the range of values within which the sample mean is expected to fall if sampling error is a reasonable explanation as to why the sample data are discrepant from the value of the mean specified in the null hypothesis. Critical values are usually specified in units of standard scores.
The conditions below are all appropriate for the use of a one-way ANOVA except____.
The dependent variable is on an ordinal scale of measurement.
SPSS provides a calculation in the output labeled as "the standard error of the mean." What is the specific information that is being provided?
The estimated standard error of the mean.
F Test
The evaluation of the F ratio formed by dividing the mean square between by the mean square within to determine if it is larger than the critical value of F.
Between-Group Variability
The extent to which mean scores in the different groups defined by the levels of the independent variable are similar.
Variability
The extent to which scores are similar to one another.
Kurtosis
The flatness or peakedness of one distribution relative to another.
Range
The highest score minus the lowest score.
In making a two dimensional graph
The horizontal dimension is the X axis, and the vertical dimension is the Y axis
Null Hypothesis
The hypothesis that is assumed to be true for purposes of conducting a statistical test. It is typically a hypothesis of no difference, no effect, or no relationship.
Null Hypothesis
The hypothesis that is assumed to be true for purposes of conducting a statistical test. It is typically a hypothesis of no difference, no effect, or no relationship. The symbol for the null hypothesis is H0.
In writing up the statistical results of a study, which of the following is most likely to appear in a research report? (Read all the answers and choose the best one.)
The hypothesis was confirmed at the .01 level, F(2)=9.90. The strength of the relationship, as indicated by eta-squared, was .45.
Levels
The levels of an independent variable are the different values that the independent variable can have. For example, for gender, there are two levels or values, male and female.
Joint Probability
The likelihood of observing each of two events.
Treatment Effect
The mean for a group minus the grand mean.
Grand Mean
The mean value on the dependent variable for all individuals in the experiment.
According to the classical view of probability theory, the probability of event A equals .____.
The number of events favoring A/the total number of possible observations
Probability
The number of outcomes favoring an event divided by the total number of possible outcomes.
Probability
The number of outcomes favoring or satisfying an event divided by the total number of possible outcomes.
relative frequencies
The number of times that a score occurs in a set of scores divided by the total number of scores.
Frequency
The number of times that a score occurs in a set of scores.
Percentile Rank
The percentage of scores in a distribution that occur at or below a given value.
Median
The point in the distribution of scores that divides the distribution into two equal parts. Fifty percent of the scores occur above the median and 50% of the scores occur below it.
Relative Standing
The position of a score relative to other scores in the distribution.
Standard Deviation
The positive square root of the variance. It is, roughly speaking, how far scores deviate, on average, from the mean. It is one of the more intuitive measures of variability. When it equals zero, there is no variability in the scores. The higher the value of the standard deviation, the more variability there is, everything else being equal.
Inferential Statistics
The practice of taking measurements on a sample and then, from these observations, inferring the value of a population parameter.
Conditional Probability
The probability of an event occurring given that some other event has occurred.
Beta
The probability of making a Type II error.
Power
The probability of not making a Type II error. It equals one minus Beta. It is the probability that an investigator will correctly reject the null hypothesis when it is false.
Probability Value (p Value)
The probability of obtaining a result as extreme or more extreme than the one observed, given that the null hypothesis is true.
Measurement
The process of translating empirical relationships between objects into numerical relationships.
Interquartile Range
The range of the scores after the top 25% of the scores and the bottom 25% of the scores have been "trimmed" or eliminated.
Deviation Scores
The raw score minus the mean; it is how far a score is from the mean.
Bivariate Relationship
The relationship between two variables.
cumulative relative frequencies
The relative frequency associated with a given score plus the sum of all relative frequencies for scores less than that score.
Non-Experimental Strategy
The same as an observational strategy.
Percentile
The score corresponding to a given percentile rank.
Mode
The score that occurs most frequently.
Standard Error of the Difference
The standard deviation of a sampling distribution of the difference between two independent means. It reflects the average amount of sampling error in a mean difference.
Standard Error of the Mean
The standard deviation of a sampling distribution of the mean.
Interval Estimation
The strategy of calculating and interpreting confidence intervals.
Mean Square Between
The sum of squares between divided by the degrees of freedom between. It is a variance estimate and reflects between-group variability.
Variance
The sum of squares divided by the sample size. It is the average squared deviation from the mean (so it is sometimes called a mean square, for short). It is an index of variability that, when equal to zero, means there is no variability. The higher the value of the variance, the more variability there is, everything else being equal.
Mean Square Within
The sum of squares within divided by the degrees of freedom within. It is a variance estimate and reflects within group variability.
Sum of Cross-Products
The sum of the product of the X scores times the Y scores.
Mean
The sum of the scores divided by the sample size. It is the arithmetic average.
Sum of Squares
The sum of the squared deviation scores. It is an index of variability that, when equal to zero, means there is no variability. The higher the value of the sum of squares, the more variability there is, everything else being equal.
Marginal Frequencies
The sums of the frequencies in the corresponding row or column of a contingency table.
Negatively Skewed
The tendency for scores to cluster above the mean.
Positively Skewed
The tendency for scores to cluster below the mean.
Skewness
The tendency for scores to cluster on one side of the mean.
At both Little Community University (LCU) and Huge State University (HSU), exactly 100 students are majoring in art. At LCU, the relative frequency of students majoring in art is 0.28 or 28%. At HSU, the relative frequency of students majoring in art is 0.01 or 1%. If exactly 100 students are majoring in art at both places, then why is the relative frequency so different?
The total number of students at HSU is much larger than the total number of students at LCU. So 100 students is a small percentage of the total at HSU, but a large percentage at LCU.
Descriptive Statistics
The use of numerical indices to describe either a population or a sample.
Confidence Limits
The values that define the range of the confidence interval. The upper limit is the larger value and the lower limit is the smaller value.
Within-Group Variability
The variability of scores within a given group.
Which of the following is NOT one of the conditions for using the one-sample z test?
The variable under investigation is qualitative and measured on an ordinal level.
ordinate
The vertical dimension of a graph; the Y axis.
Dr. Hind used a One-Way ANOVA, between subjects design and has set his alpha level at .05 nondirectional. He reported the following: F(2, 19)=6.99. Based on this information what can you conclude?
There is a significant main effect of your independent variable
Dr. Hind used a One-Way ANOVA, between subjects design and has set his alpha level at .05 nondirectional. He reported the following: F(2, 19)=6.99. Based on this information what can you conclude?
There is a significant main effect of your independent variable.
Why are standard scores important?
They allow us to compare scores from different distributions.
You have conducted a study using a therapy technique to reduce depression. Out of 10 clients, you found a reduction for 6 of them. Can we conclude that the therapy technique works?
This cannot be determined until a z-score is calculated and compared with an alpha level criterion.
The problem with testing each pair of group means by conducting three independent groups t tests is that multiple t tests increase the probability of making a _____ for at least one of the tests beyond the probability specified by the alpha level.
Type I error
A researcher performs a hypothesis test and rejects the null hypothesis. If, in reality, the null hypothesis should not have been rejected, then the researcher has made a:
Type I error.
A statistical test led Dr. Jones to incorrectly decide that his independent variable had no affect on the dependent variable, when in fact it did. Dr. Jones committed a(n)
Type II error
Nominal Measurement
Type of measurement in which numbers are merely used as labels.
Ordinal Measurement
Type of measurement in which the assignment of numbers permit one to order objects or people along a continuum or dimension.
Interval Measurement
Type of measurement in which the assignment of numbers permit one to order objects or people along a continuum or dimension. In addition, numerically equal distances on the number scale correspond to equal amounts on the underlying dimension.
Ratio Measurement
Type of measurement that has all the properties of interval measures but which also map onto the underlying dimension in such a way that ratios between the numbers represent ratios of the dimension being measured.
Degrees of Freedom
Used to indicate the number of pieces of information that are "free of each other" in the sense that they cannot be deduced from one another.
Counterbalancing
Using more than one sequence of conditions or levels of the independent variable so as to evaluate the effects of order of condition.
Rejection Region
Values of a test statistic that are outside the range of critical values. If the critical values are -1.96 to 1.96, then the rejection region refers to all values greater than 1.96 and less than -1.96.
Explained Variance
Variability in the dependent variable that is associated with (explained by) the independent variable.
Unexplained Variance
Variability in the dependent variable that is due to factors other than the independent variable.
Qualitative Variables
Variables measured on a nominal level.
Quantitative Variables
Variables measured on ordinal, interval, or ratio levels.
Discrete Variables
Variables that can assume only a finite number of values or that have a finite number of values that can occur between any two points.
Continuous Variables
Variables that have an infinite number of values between any two points.
Bimodal
When a distribution has two modes.
Experimental Strategy
When a set of procedures or manipulations is performed to create different values of the independent variable for the research participants.
A developmental psychologist's is studying at what age children develop a fear of strangers. She studies infants at 2 months old (n=8) and than again at 6 months old (n=8) and measures how fearful they are to strangers. Using a scale of (0=no fear to 5=very afraid). What type of experimental design is she using?
Within-subjects design
The general formula of the linear model (the regression line) is _____.
Y = a + bX
JJ just used an Independent groups t-test to analyze his data for his Master's degree and found that the t-value or his calculated t was equal to 3.05 based on his degrees of freedom of 60. The table t was 1.671 and he set his alpha level at .05 directional. Can he reject the null hypothesis and conclude the groups are significantly different?
Yes, he can reject the null.
In an APA results section and independent groups t-test is written as follows t(82) = 5.66, p<0.05. Based on this information were the groups significantly different?
Yes, since p<0.05.
Dr. Johnson is studying the effects of aspirin on heart rate. He believes that aspirin will lower heart rate. He has 2 groups, the treatment group receives 80mg of aspirin for a week (n=4), the control group received a placebo (n=4), and their heart rates are compared. Below is his data in which heart rate is beats per minute. He sets his alpha level at 0.05 and uses a directional test. Can you reject the null hypothesis and conclude that aspirin lowers heart rate?
Yes, since the calculated t is greater than the table t.
Eta-squared is _____ estimator in that it tends to slightly _____ the strength of the relationship in the population across random samples.
a biased; overestimate
cumulative percentages
a cumulative relative frequency times 100
If the variability of scores in the populations is held constant, then increasing sample sizes results in _____ in the standard error of the difference.
a decrease
In order to determine the nature of the relationship with a one-way ANOVA:
a multiple comparison procedure or post-hoc test must be used.
Within subjects designs have several advantages, which include_____. (Read all the answers and choose the best one.)
a smaller sample size is needed AND has fewer disturbance variables
Probability theory is important to understand because
a. probability is the foundation of inferential statistics used in psychology. b. probability is the foundation of many descriptive statistical techniques used in psychology. c. understanding basic probability theory helps in understanding many statistical tests.
When scores on Y are predicted from scores on X using a regression equation, the amount of error for a given individual can be represented by the discrepancy between that person's:
actual and predicted Y scores
In a truly random sample
all members of the population have an equal chance of being selected
In a truly random sample from the population:
all members of the population have an equal chance of being selected.
In order to select the correct statistical test, you need to know:
all of these // whether the dependent variable at least approximates the characteristics of an interval scale b. the number of levels of the independent variable c. whether the design is between-subjects or within-subjects d. whether the variables are quantitative or qualitative
Independent-groups t tests may be used to analyze the relationship between two variables when _____.
all of these// the independent variable is between-subjects in nature the dependent variable is quantitative the independent variable has only two levels
A common strategy to control confounding variables is (are):
all of these// random assignment to experimental groups. matching. holding a variable constant.
In hypothesis-testing, the size of the portion of the tails of the sampling distribution that fall outside your critical values (i.e., your "rejection regions") is determined by your ________.
alpha level
Rejection regions are determined with reference to a probability value known as the ____ :
alpha level
If a researcher were to subtract the mean of a distribution from each score in that distribution and add the differences, the resulting value would _____.
always be zero
If a researcher were to subtract the mean of a distribution from each score in that distribution and add the differences, the resulting value would ________.
always be zero
If a confounding variable has been included in the design of an experiment, then interpretation of the results would be ___.
ambiguous
An advantage of using a within-subjects design instead of a between-subjects design is that within-subjects designs
are more economical in terms of participants and reduce confounding variables.
When there is an even number of scores, the median is the ________.
arithmetic average of the two middle scores
The main difference between parametric and nonparametric statistics is that parametric statistics
assumes that the population distribution is normally distributed
You discovered that your Statistics professor never gives out raw scores only standard scores. You are relieved because you saw a "0" on the top of your paper. This means you scored ________
at the mean
The variance is the:
average of the squared deviations from the mean.
As your sample size gets smaller (especially as it gets below about 30-40), the t-distribution and the z-distribution:
become more and more different.
Other things being equal, as the population standard deviation becomes larger, the standard error of the mean
becomes larger
The more different the group means are from each other (as defined by the levels of the independent variable), the more _____ there is.
between-group variability
Scales of measurement used in behavioral research include all of the following except
binocular
A correlation coefficient of _____ means the two variables form a perfect linear relationship. (Choose the best answer.)
both -1.00 and +1.00
In a frequency histogram, the horizontal dimension is called the ________, and the vertical dimension is called the ________.
both X axis; Y axis / abscissa; ordinate
A 95% confidence interval for the mean:
both defines the range of scores in which there is a .95 probability the population mean will fall, and is computed using the standard error of the mean.
In a one-way ANOVA between groups, variability include(s) ________.
both individual differences and effect of the independent variable
The difference between the sample variance and the variance estimate is that the variance estimate
both is computed with N-1 in the denominator, and is computed with the degrees of freedom in the denominator.
Confidence intervals are used to
both provide information about sampling error, and determine whether or not a relationship exists between the independent and dependent variables.
The F value for a one-way between subjects ANOVA is appropriate when the dependent variable is ________ in nature and measured on a level that approximates ________ level of measurement.
both quantitative; interval / quantitative; ratio
The t-distribution should be used when ________, while the z-distribution should be used when ________.
both sample sizes are small; and are known / sample sizes are small; sample sizes are large
In a one-way ANOVA, between groups variability includes_____.
both sampling error AND the effect of the independent variable
In a One-way analysis of variance, what factors influence within group variability?
both sampling error and individual differences
In a One-way analysis of variance, what factors influence between group variability?
both sampling error and treatment effects
What measures are used to determine the amount of variability that occurs in a data set?
both variance and standard deviation
The calculation of eta-squared _____ when the statistical test is nonsignificant.
can be useful
One potential problem with within-subjects designs is the fact that the treatment in the first condition may have _____ effects that influence performance in the second condition.
carry-over
A matched-subjects design often has the power of a within-subjects design while avoiding the problem(s) of ___.
carry-over effects
An important methodological problem to consider in a correlated-groups design is ______.
carry-over effects of the independent variable
Imagine you have five questionnaires you want to hand out to a group of people. What is the counting rule used to determine how many sets of five you have when ordering of the questionnaires does not matter?
combinations
A _____ indicates the likelihood that an event will occur given that some other event occurs.
conditional probability
An investigator was interested in determining if the size of a coin influenced judgments of numerosity when people were asked to make judgments of how many coins were in a pile of coins. Fifteen subjects were asked to estimate the number of pennies in a display of 43 coins. In another condition, 15 different subjects estimated the number of dimes for the same display. In the above experiment, the size of the coin is also related to its value. Value is thus a ____.
confounding variable
Counterbalancing is a method to control for
confounding variables.
A sample distribution:
contains raw scores from a sample.
A sampling distribution:
contains sample means, and will look roughly normal as long as sample sizes are large enough only.
What type of data table is useful for calculating probabilities?
contingency table
The true population that may be infinite is considered a _____ variable.
continuous
Variables that have an infinite number of values between any two points called:
continuous
The mean, median, and mode are applicable to:
continuous and discrete variables
Dr. August is investigating whether or not women have more job satisfaction after they are given praise or a gift card (different levels of rewards). She has recruited 10 women from Intel and for 6 months the women received praise then the next 6 months they received a gift card. The conditions were counterbalanced and a survey on job satisfaction was distributed following the treatment conditions or every 6 months until the research was completed. After finishing her study Dr. August was busy writing a grant, therefore she has asked you to analyze her data below. Please note the survey was based on 1-5 Likert scale with high scores indicating high job satisfaction and low score indicating dissatisfaction. What statistical test should be used to analyze Dr. Augusts' data?
correlated-groups t test
Dr. Hinson wants to investigate if rats fed a diet enriched with omega will do better at a maze learning task than when these same rats are given a normal diet. He has a total of 30 rats for his experiment. What type of statistical test should Dr. Hinson use in analyzing his data?
correlated-groups t test
Dr. Miguel is studying stages of verbal ability in kids. He test the kids on verbal ability at age 5 years then again at age 7. What type of statistical test should he use to analyze his data?
correlated-groups t test
Dr. Miguel is studying stages of verbal ability in kids. He test the kids on verbal ability at age 5 years then again at age of 7. What type of statistical test should he use to analyze his data?
correlated-groups t test
If a researcher used a matched-subjects design, which of the following statistical tests should be used?
correlated-groups t test
Comparing the relationship happiness and owning a pet is _____.
correlational
The procedure of _____ may be used in order to evenly distribute carry-over effects across conditions.
counterbalancing
Jacob is studying the development of motor coordination in monkeys from birth to old age for his Master's thesis. However, he only has one year to collect his data. What type of design should he use?
cross-sectional
As the variability of scores in the population decreases, the standard error of the mean _____, other things being equal.
decreases
The Pearson correlation coefficient (r) provides information about the ________.
degree of a linear relationship between two variables
In statistics, the phrase _____ is used to indicate the number of pieces of information that are "free of each other" in the sense that they cannot be deduced from one another.
degrees of freedom
In statistics, the phrase ________ refers to the number of pieces of information that are "independent of each other" in that they cannot be deduced from one another.
degrees of freedom
In general, estimation of a population parameter becomes more accurate as ____.
degrees of freedom (df) of the sample increases
Correlation and regression both can help researchers to
describe a linear relationship between two variables
Correlation and regression both can help researchers to:
describe a linear relationship between two variables.
Measures of Central tendency are an example of ________.
descriptive statistics
When the Professor tells the class the mean on the exam. She/he is using ________.
descriptive statistics
In a one-sample z test, when you have evidence or reasoning to suggest that your sample mean will most likely differ from the hypothesized population value in one direction (for example, "it will be higher than the hypothesized value"), you would use a:
directional ("one-tailed") test.
An alternative hypothesis that specifies that a population mean is different from a given value and also indicates the direction of that difference is called a:
directional alternative hypothesis
The number of people in your family is considered a ________ variable.
discrete
A variable that influences the dependent variable but is unrelated to the independent variable is called a ____ variable.
disturbance
The one-sample z test is used to determine whether the difference between your sample mean and the hypothesized population value is:
due to sampling error.
The Tukey HSD test discerns the nature of the relationship by testing a null hypothesis for:
each possible pair of group means
According to the central limit theorem, if you selected every possible random sample of size N from some population and computed a mean (bar over X) for each of these samples, then the average (i.e., mean) of these
equal to the population mean
According to the central limit theorem, if you selected every possible random sample of size N from some population and computed a mean (mc086-1.jpg) for each of these samples, then the average (i.e., mean) of these sample means will always be ____.
equal to the population mean
A t score is analogous to a z score except that it represents the number of _____ that a sample mean is from the hypothesized value of mu (the population mean).
estimated standard errors
Dr. August is investigating whether or not women have more job satisfaction after they are given praise or a gift card (different levels of rewards). She has recruited 10 women from Intel and for 6 months the women received praise then the next 6 months they received a gift card. The conditions were counterbalanced and a survey on job satisfaction was distributed following the treatment conditions or every 6 months until the research was completed. After finishing her study Dr. August was busy writing a grant, therefore she has asked you to analyze her data below. Please note the survey was based on 1-5 Likert scale with high scores indicating high job satisfaction and low score indicating dissatisfaction. What should be done after the inference about the relationship has been determined?
eta-squared should be calculated.
Which of the following indices measures the strength of the relationship between the independent and dependent variables?
eta2
The nature of the relationship is determined by ________.
examining the sample means
Assuming the null hypothesis is true, we can specify an _____ of an investigation.
expected result
Comparing how well males and females perform on a video game that simulates landing a space shuttle is _____.
experimental
An r value of .20 always indicates a small, weak, unimportant correlation between two variables in the behavioral science literature.
false
An advantage of matching over holding a variable constant is that
for each individual in one group there is a comparable individual in each of the other groups who has the same value on the confounding variable.
If the observed value of F is_____than the critical F, then the null hypothesis is rejected. We can conclude that the independent variable ____ influenced the dependent variable.
greater; has
If the observed value of F is ________ than the critical F then the null hypothesis is rejected and we can conclude that ________ an influence of the independent variable.
greater; there is
If you compute a negative Pearson r between two variables (X and Y), this tells you that ________.
higher scores on X are associated with lower scores on Y
A common strategy to control disturbance variables is (are):
holding a variable constant.
In practice, it is _____ to compute the exact amount of sampling error that occurs.
impossible
In the case of a negative relationship, as scores on X _____, scores on Y _____.
increase; decrease
As the degrees of freedom associated with an estimate _____, the accuracy of the estimate tends to _____.
increase; increase
The effect of performing multiple t tests on all pairs of groups after finding a significant F ____.
increases the probability of making at least one Type I error
The experimenter will manipulate the ________
independent variable
The x-axis is reserved for the ________ variable and the y-axis is where the ________ variable is found.
independent; dependent
Dr. Johns is studying the effects of caffeine on memory. After comparing the subjects taking a placebo with those taking caffeine pills he concludes that caffeine enhances memory. He most likely used ________ to determine the difference between groups.
inferential statistics
In most behavior science research a goal is to describe very large groups of people. This would be considered a(n) ________.
infinite population
Estimating population parameters can be accomplished with reference to _____ populations.
infinite, small, large
You are taking a survey where they ask you to rank questions as follows: 1= strongly agree 2= agree 3= neutral 4= disagree 5= strong disagree This is what type of scale of measurement?
interval
A combination of a set of objects or events is a sequence in which the internal ordering of elements is:
irrelevent
A treatment effect
is found by comparing the mean score in each condition with the grand mean
A problem with holding a variable, such as gender, constant is that
it may reduce the generalizability of the results.
The defining characteristic of the regression line is that:
it simultaneously minimizes the distances of all data points from the line.
The critical t value needed to reject the null will always be _____ than a Z value for the same set of scores.
larger
The symbol for true population is ________.
line with u
Research in the behavioral sciences is often concerned with _____ relationships.
linear
The extent of _____ between two variables is indexed by a statistic known as the Pearson correlation coefficient.
linear approximation
In a grouped frequency distribution, the conventional starting point for the lowest interval is the closest number evenly divisible by the interval size that is equal to or less than the ________.
lowest score
When we use inferential statistics to address a research question, we:
make generalizations from a sample to the population it was sampled from.
If the sample size is less than 15, the correlated groups t test may show inflated Type I errors for data that are:
markedly skewed
The _____ is simply the arithmetic average of a set of scores.
mean
Which measure of central tendency is affected by extreme scores?
mean
How would you describe the relationship between the mean, median, and the mode in a positively skewed distribution?
mean > median > mode
The sum of squared deviations from the ________ will always be ________ the sum of squared deviations around any other value.
mean; less than
In APA journal articles researchers usually report the _____and_____of the data set.
mean; standard deviation
When presenting measures of central tendency and variability, most researchers report _____ and_____ , which are the most _____ descriptive statistics.
means; standard deviations; frequently used
Scales of measurement used in Behavioral research includes all of the following except
molecular
Assume a researcher has grouped people into categories by age, as follows: 20-29, 30-39, 40-49, 50-59. A person can only be placed into one of these categories, thus these categories are said to be
mutually exclusive
Assume a researcher has grouped people into categories by age, as follows: 20-29, 30-39, 40-49, 50-59. A person can only be placed into one of these categories, thus these categories are said to be:
mutually exclusive
Factorial
n!=(n)(n-1)(n-2)(n-3)...(1)
The degrees of freedom for the independent groups t test are equal to:
n1 + n2 - 2
Which type of relationship is graphed in the following scatterplot?
negative linear relationship
In principle, we can _____ accept the null hypothesis as being _____ via our statistical methods.
never; true
In ANOVA, the F test is _____.
nondirectional
When the independent variable has more than two levels which of the following statistical techniques may be used to investigate the relationship between the independent and dependent variables?
none of these
If you have a small sample size what type of statistics should you use?
nonparametric statistics
According to the central limit theorem, the sampling distribution of the mean approximates a _____ distribution as the sample size _____.
normal; approaches infinity
If the observed t value exceeds the critical t value, then the difference between the sample means is most likely ____.
not attributable to sampling error
Assume that you wanted to test the effectiveness of a certain therapy technique in reducing depression among clients. According to statistical hypothesis testing the hypothesis that the technique's effectiveness will be equal to chance is called the _____.
null hypothesis
Assume that you wanted to test the effectiveness of a certain therapy technique in reducing depression among clients. According to statistical hypothesis testing the hypothesis that the technique's effectiveness will be equal to chance is called the ________.
null hypothesis
Informally, the _____ hypothesis can be thought of as the hypothesis of _____.
null; no difference
Directional tests are often referred to as :
one-tailed tests
Dr. Wright is studying the effects of 3 different doses of antihistamines on drowsiness in a total of 45 subjects. One group will receive 10mg, the second group 20mg, and the last group 40mg. They were then asked to rate how drowsy they felt on a scale of 1-10. What statistical test should Dr. Wright use?
one-way ANOVA between subjects design
In the absence of any other information, the sample mean that one observes is _____ of the value of the population mean.
ones "best guess"
Dr. Hinson wants to investigate if rats fed a diet enriched with omega will do better at a maze learning task than when these same rats are given a normal diet. He has a total of 30 rats for his experiment. If Dr. Hinson fails to implement the appropriate changes to his experiment a potential confound may be ________
order effects
Imagine that you recently won 1st place in a local talent contest. This is an example of aNo ________ scale of measurement.
ordinal
Sally's dog just received second place in obedience class. This is considered a(n) ________ scale of measurement.
ordinal
In frequency histograms, the frequencies associated with score values are labeled along the ________, and the score values are labeled along the ________.
ordinate; abscissa
An unbiased estimator of a population parameter is one whose average (mean) over all possible random samples of a given size equals the value of the:
parameter in question
The entire population is considered a _____ but a sample of the population is a _____.
parameter; statistic
A counting rule that involves a set of objects of events in an ordered sequence is called a ____.
permutation
A distribution that has a mean of 10 SD of 5 is much more ________ than a distribution with a mean of 10 and SD of 2.
platykurtic
If a distribution is less peaked than another, it is said to be more ________; if it is more peaked than another, it is said to be more ________.
platykurtic; leptokurtic
Use of Greek notation implies that the parameters describe ____ characteristics.
population
When using the correlated groups t test, it is assumed that the _____ is normally distributed.
population of difference scores
Which of the following is NOT true about probability theory?
probability theory is of little use to science or psychology.
Relative to the Scheffè procedure, the Tukey HSD test ___.
produces fewer Type II errors
When a correlation is perfect and positive, which of the following holds true?
r=1.00
The _____ is the highest score minus the lowest score.
range
Dr. Johnson hypothesizes that NutraSweet reduces short-term memory. In order to test her hypothesis she gives 10 subjects a beverage containing NutraSweet (treatment group) and 10 subjects a beverage without NutraSweet (control group). She then measures the subject's ability to recall a list of numbers. Dr. Johnson's dependent variable is measured using a(n) ________ scale of measurement.
ratio
If our observations are so discrepant from the expected result that the difference cannot be attributed to chance, we will _____ the null hypothesis.
reject
The number of times that a score occurs divided by the total number of scores is termed a(n)________.
relative frequency
_____ refers to the extent to which conclusions drawn on the basis of a statistical test (for example, rejection of the null hypothesis) are unaffected by violations of the assumptions underlying the test.
robustness
The central limit theorem states that a sampling distribution of the mean for a population will approximate that of a normal distribution as ____. (Be sure to choose the BEST answer.)
sample size increases
Outliers are most likely to raise interpretational complexities when:
sample sizes are small
Outliers are most likely to produce misleading results when
sample sizes are small.
A _____ can be formally defined as a theoretical distribution consisting of the mean scores for all possible random samples of a given size that could be drawn from a population.
sampling distribution of the mean
Greater variability of scores within a group is indicative of greater variability of scores within the corresponding population and, thus, a greater amount of:
sampling error
In an ANOVA context, within-group variability reflects _____.
sampling error
The fact that a sample statistic may not equal the value of its corresponding population parameter is said to be the result of:
sampling error
The independent samples t test is reasonably robust to violations against which assumptions? (Read all answers and choose the best one.)
scores in each population are normally distributed AND scores in the two populations have equal variance
If the standard error of the mean is small, then all the sample means based on a given sample size (N) will tend to be _____, and all will tend to be _____ the population mean.
similar ; close to
When there is an odd number of scores, the median is:
simply the middle score
The __________ indicates the number of units variable Y changes as variable X changes by 1 unit.
slope
If the effects of individual differences have not been removed from the raw scores, the eta2 for the correlated-groups case would be _____ than if they have been removed, everything else being equal.
smaller
The fact that there is a positive relationship between the murder rate and ice cream consumption can be explained by the fact that both of these increase during hot weather. This is an example of a _____.
spurious relationship between the murder rate and ice cream consumption
The fact that there is a positive relationship between the murder rate and ice cream consumption can be explained by the fact that both of these increase during hot weather. This is an example of a:
spurious relationship between the murder rate and ice cream consumption.
The amount of sampling error is determined by ________.
standard error of the difference
The number of standard deviation units that a score falls above or below the mean is represented by the ________.
standard score
In the study of traditionalism and ideal family size, a correlation of .66 was found indicating a ______.
strong positive relationship
if a researcher sets a critical z value equal to ±1.96, then test statistics falling beyond that range _____.
suggest that the alternative hypothesis is true
A mean square is simply a _____ divided by its corresponding degrees of freedom
sum of squares
SS is the abbreviation for _____, which is shorthand for _____.
sum of squares ... sum of squared deviations from the mean
Based on the Method of Presentation section of Chapter 8, suppose you obtained a statistically significant t value of 2.5 with 15 subjects using an alpha level of .05. How would this be reported in a psychological journal?
t (14) = 2.5, p < .05
Based on the Method of Presentation section of Chapter 11, suppose a researcher is reporting the results of a correlated groups t test. If she obtained a t value of 2.17 with 10 subjects in each group using an alpha level of .05 and a nondirectional test, how would she report this in a psychological journal?
t (9) = 2.17, p > .05
Dr. August is investigating whether or not women have more job satisfaction after they are given praise or a gift card (different levels of rewards). She has recruited 10 women from Intel and for 6 months the women received praise then the next 6 months they received a gift card. The conditions were counterbalanced and a survey on job satisfaction was distributed following the treatment conditions or every 6 months until the research was completed. After finishing her study Dr. August was busy writing a grant, therefore she has asked you to analyze her data below. Please note the survey was based on 1-5 Likert scale with high scores indicating high job satisfaction and low score indicating dissatisfaction. In Dr. Augusts' result section how should she report the results of the t-test you calculated for her?
t(4) = 6.00, p<0.05
An example of an omnibus test is
the F ratio
The standard error of estimate is an indication of:
the average error resulting from predicting Y values from X scores using a regression equation.
John waited an average of 5 minutes at a local Starbucks for his coffee with a SD of 2.00. On campus he waits 4 minutes for coffee with a SD of 3.00. What coffee vendor has more variability in its distribution of coffee?
the campus coffee house
The main difference between a correlated groups t-test and Independent groups t-test include(s)_____.
the correlated groups t-test assumes the independent variable is within subjects
The main difference between a correlated groups t-test and Independent groups t-test include(s):
the correlated groups t-test assumes the independent variable is within subjects.
The conditions below are all appropriate for the use of a one-way ANOVA except:
the dependent variable is on an ordinal or quantitative scale of measurement.
The Independent groups t-test assumes all of the following except ________
the dependent variable is ordinal
Which of the following conditions would preclude the use of a correlated-groups design?
the experimental manipulation has known carry-over effects
abscissa
the horizontal dimension of a graph; X axis
In the use of the binomial expression for hypothesis testing, the initial assumption is that the frequency of the event will be equal to:
the likelihood that the event will occur by chance.
The sampling distribution of the mean contains
the means of all possible random samples of a given size from a population.
If a researcher randomly places subjects in a high, medium, or no stress condition, then the control condition is _____.
the no stress condition
A researcher in sports psychology is interested in the resting pulse rate of long-distance college runners. She does some research and finds data from the U.S. Olympics team showing their long-distance runners have an average resting pulse rate of 60 beats per minute. She wants to know if the pulse rate for college runners is different from the Olympic team's mean of 60. She randomly samples 20 long-distance runners, and finds mc089-1.jpg What is the appropriate test statistic for this data?
the one-sample t test
A correlated groups t-test is a parametric test and assumes_____.
the population is normally distributed
A correlated groups t-test is a parametric test and assumes:
the population is normally distributed.
The mean of the sampling distribution of the mean is always equal to
the population mean.
A scatterplot primarily is intended to show
the relationship between X and Y scores.
A one-sample t test is used in situations where
the s must be estimated from a sample.
The formula for computing eta-squared for the correlated groups t test is _____ that for the independent groups t test.
the same as
Based on the Method of Presentation section of Chapter 3, the symbol n indicates:
the sample size in a particular group.
The correlated-groups t test is robust to violations of the normality assumption if _____.
the sample size is relatively large
In the Applications to the Analysis of Social Problems section of Chapter 8, the confidence intervals for the mean age at which mothers think parents should talk with children about sex and the confidence intervals for the mean age at which mothers think parents should talk with children about birth control were relatively narrow. What does this mean?
there are small amounts of sampling error
The major problem with the mode as a measure of central tendency is that:
there can be more than one modal score.
Researchers were recently surprised to learn that the correlation between the type of clothing people wear when they are not working and the type of jobs they hold is approximately zero. Based on this, we can conclude that _____.
there is no relationship between the types of clothing worn in non-work settings and the type of work that people do
Researchers were recently surprised to learn that the correlation between the type of clothing people wear when they are not working and the type of jobs they hold is approximately zero. Based on this, we can conclude that ________.
there is no relationship between the types of clothing worn in non-work settings and the type of work that people do
According to classical probability theory, the act of selecting a card at random from a deck is called a(n):
trial
A correlated groups t test is analogous to an independent groups t test with the effects of individual differences extracted from the dependent variable.
true
Fractions may be included when reporting the means for discrete variables.
true
In deciding how many groups to report, a researcher must strike a balance between having so many that he/she cannot easily comprehend the patterns of the data and having so few that the table lacks precision.
true
In a binomial probability calculation, there are ____ possible outcomes.
two
The F test is relatively robust to violations of the homogeneity of variance assumption when the population variance of one group is as much as _____ larger than the population variance of the other groups.
two to three times
If the sample size of a study is small and a high value of eta squared is observed, but there is no statistically significant effect of the independent variable on the dependent variable, we may conclude that ______.
unless a larger sample is used we can't be certain about the effects of the independent variable
The 95% confidence intervals are ________ reported in APA results sections.
usually
Although the correlated t test is usually a more powerful statistical test than the independent groups t test, one exception to this is
when r is so close to 0 that the magnitude of estimated standard errors is comparable in the two tests.
Dr. Johnson hypothesizes that nutrasweet reduces short-term memory. In order to test her hypothesis she gives 10 subjects a beverage containing nutrasweet (treatment group) and 10 subjects a beverage without nutrasweet (control group). She then measures the subject's ability to recall a list of numbers. In this experiment, the independent variable is_______and the dependent variable is______.
whether the subjects received nutrasweet: recall for a list of numbers
As the alpha level becomes larger, the power of a statistical test ____.
will increase
If two variables are related in a fashion that is nonlinear, Pearson correlation:
will not be sensitive to this
Sampling from a finite population _____ is analogous to sampling from an infinite population _____.
with replacement; without replacement
In an APA results section and independent groups t-test is written as follows t(82) = 5.66, p<0.05. Based on this information were the groups significantly different?
yes, since p<0.05
In a one-sample z-test, the term "statistically significant difference" tells you:
your sample mean differs from the hypothesized value by more than would be expected from sampling error alone.
A physical therapist is interested in whether people with different bone structures will be affected differently by her new therapy program. She randomly chooses two groups of patients based on bone type: small-boned frame (XS) and large-boned frame (XL). Patients all had similar injuries. She measures the number of days each person takes to progress to a predetermined point. Here is the data: Small Bones Large Bones Mean 10.00 12.00 n 15 12 s2 4.5 5.0 SS 108 105 What is the null hypothesis?
µS = µL