Stats ch13, 15-17

Identify a purpose of discriminant function analysis. a) simplifying the calculation of a multiple correlation coefficient b) correlating predictor variables and a categorical criterion variable c) determining if many variables can be described by a few factors d) reducing the number of variables to a more manageable level

B

Identify an advantage of closed-ended questions in a survey. a) They are easier to construct. b) They are easier and faster to tabulate. c) They permit follow-up by the interviewer. d) They allow more freedom of response.

B

In a causal-comparative study, groups should be compared on: a) all variables that are easily measured. b) those variables suggested by the researcher's rationale. c) only quantitative variables. d) only the variable of primary interest.

B

In which of the following scenarios would one most likely find a correlation coefficient of zero or close to zero? a) age of cars correlated with their trade-in values b) shoe sizes of adult males correlated with their salaries c) intelligence correlated with grades in arithmetic d) age correlated with the cost of life insurance

B

Partial correlation is a procedure for: a) simplifying the determination of a regression line. b) controlling a subject characteristics threat. c) reducing the number of variables to a more manageable level. d) exploring theoretical cause-effect relationships.

B

Path analysis is a technique for: a) determining a regression line. b) testing the likelihood of a causal connection among three or more variables. c) estimating the amount of error associated with a predicted score. d) reducing the effects of an extraneous variable.

B

A regression line is used to: a) connect the points in a scatterplot. b) correct data for the regression threat to internal validity. c) estimate the score for an individual on one variable based on the individual's score on another variable. d) estimate the extent to which material has been forgotten.

C

A researcher would be likely to use matching in a randomized posttest-only control group design: a) in order to control for a possible history threat to internal validity. b) in order to control for a loss of subjects. c) in an attempt to increase the likelihood that the groups of subjects in an experiment will be equivalent. d) only if he or she misunderstood randomization.

C

In a ________, different samples from a population whose members may change are surveyed at different points in time. a) cohort study b) panel study c) trend study d) census

C

In the context of controlling extraneous variables, ________ is used to equate groups statistically on the basis of a pretest or other variables. The posttest scores of the subjects in each group are then adjusted accordingly. a) matching b) holding certain variables constant c) analysis of covariance d) randomization

C

Interviewers should be trained to do all of the following EXCEPT: a) establish a friendly relationship with the respondent. b) follow up on ambiguous answers. c) show their own views regarding the study. d) ask questions in a standard manner.

C

Multiple regression is a technique for: a) eliminating the effect of extraneous variables in a correlational study. b) reviving childhood memories. c) determining a correlation between a criterion variable and the best combination of two or more predictor variables. d) adjusting scores for a data collector threat.

C

One way of providing some control in a causal-comparative research design is to: a) do a pilot study before carrying out the main research. b) obtain measurements on an additional dependent variable. c) match the subjects from comparison groups on extraneous variables. d) list all of the subjects from the population under study and randomly assign them to experimental and control groups.

C

The basic causal-comparative approach is to begin with: a) the assignment of subjects to treatment groups. b) the measurement of the independent variable. c) a noted difference between two groups and to look for possible causes for, or consequences of, this difference. d) the measurement of the dependent variable.

C

The difference between a trend study and a cohort study is that a cohort study obtains data from: a) different samples from a population whose members change somewhat. b) an entire population at different points in time. c) different samples from the same specific population. d) the same sample at different points in time.

C

The major threat to internal validity in a causal-comparative study is the possibility of a: a) maturation threat. b) regression threat. c) subject characteristics threat. d) history threat.

C

Which of the following is an example of a cross-sectional survey? a) A group of gifted children is assessed every other year until adulthood. b) An opinionnaire is mailed to the same sample of graduates every year. c) An attitude scale is given to samples of 8-, 10-, and 12-year-olds. d) None of these answers are correct.

C

________ is a technique to reduce the number of variables by searching for "clusters" of variables, all of which are correlated with each other. a) Discriminant function analysis b) Multiple regression c) Factor analysis d) Path analysis

C

Causal-comparative research is interested in exploring the cause or consequences of differences that already exist among groups of individuals. True False

True

Random replacement is a faulty procedure commonly used to handle nonresponse in telephone surveys. True False

True

The most commonly used statistical inference test in causal-comparative studies is a t-test for differences between means. True False

True

An important difference between experimental and nonexperimental research is that the independent variable in experimental research is: a) constant. b) manipulated. c) quantitative. d) randomly assigned.

B

Correlational research differs from experimental research in that: a) random sampling is not used. b) there is no manipulation of variables. c) researcher bias is not a problem. d) there is no concern regarding internal validity.

B

Experimental research differs from all other types of research in that experimental research: a) involves the making of exact measurements. b) involves the manipulation of the independent variable. c) consists of a repetition in observations. d) is highly descriptive.

B

Experimental research is the most powerful research method for: a) testing hypotheses about cause-and-effect relationships. b) identifying possible relationships. c) generalizing the results of a study. d) identifying important dependent variables.

A

In a large high school, the 25 students who scored highest on an English pretest were placed in a special class. At the end of one semester, they were given another English achievement test. This is an example of a: a) one-group pretest-posttest design. b) time series design. c) randomized control group pretest-posttest design. d) static group comparison design.

A

In a panel study, the researcher obtains information from: a) the same sample at different points in time. b) different samples at different times. c) an entire population at different points in time. d) different samples at the same time.

A

Is it important to follow up on those persons in a survey who did not return questionnaires mailed to them? a) Yes, because the sample that has already returned the questionnaire may be biased. b) Yes, because descriptive statistics require a large sample. c) No, because high precision is not needed in questionnaire surveys. d) No, because such delayed returns would be of doubtful value.

A

The purpose of most surveys is to: a) describe the characteristics of a population. b) explore cause and effect. c) obtain information from a sample. d) test hypotheses.

A

Which of the following is NOT a serious threat to the internal validity of a one-group pretest-posttest design? a) introduction of a comparison group b) history c) maturation d) statistical regression

A

Why are control groups necessary in certain experiments? a) to make it easier to estimate the influence of the independent variable b) to enable the researcher to identify the dependent variable c) to enable the researcher to repeat the experiment d) to make sure that the experimenter is honest

A

Identify a major characteristic of a survey. a) It mandates personally meeting the subjects of the survey. b) Information is collected through asking questions. c) The population is almost always studied as a whole. d) It involves the manipulation of an independent variable.

B

A notable problem with experimental research in education is that: a) there is no way to control data collector bias. b) the researcher seldom has sufficient control over treatments. c) there is no way to control a "history" threat. d) there is no way to control the subject characteristics threat.

B

A causal-comparative study is used as an alternative when experimentation is: a) impossible. b) unethical. c) too expensive. d) All of these answers are correct.

D

A comparison of delinquents vs. nondelinquents might be made in order to: a) determine probable factors leading to delinquency. b) test a theory on the development of delinquency. c) predict future behavior of delinquents. d) All of these answers are correct.

D

A correlation coefficient of −0.79 between time spent in typing practice and the number of typing errors means that: a) there is a definite cause-and-effect relationship between typing practice and typing errors. b) the correlation coefficient has no meaning as it is negative. c) there is a positive correlation between the number of hours spent in typing practice and the number of typing errors. d) the number of typing errors is likely to be low if the time spent in practice is high.

D

A factorial design can be used to: a) assess the interactions between variables. b) assess the separate effect of each independent variable. c) study the combined effect of two or more variables. d) All of these answers are correct.

D

Correlational research is used to: a) explore possible cause-and-effect sequences. b) suggest experimental studies. c) predict future behavior. d) All of these answers are correct.

D

Which of the following is NOT possible in causal-comparative research? a) measurement b) observation c) the use of comparison groups d) random assignment to treatment or comparison groups

D

Causal-comparative research establishes cause and effect. True False

False