Deck 9: Internal Validity

Question 1

Which of the following is the best way to control for the unwanted influence of maturation processes in an experiment?

A) Increase the sample size for the experiment
B) Use a pretest as well as a posttest
C) Use a control group
D) Use subjects who have stabilized with regard to maturational level

Answer 1

Use a control group

Question 2

Students with low, average, and high scores on a commercial reading achievement test were used in a study designed to determine the effectiveness of a new method of teaching fifth-grade reading. There were no other control or comparison groups. It was found that the low group made the greatest gain over a six-month period. Concluding that the new method was most effective with the low group fails to consider some of the possible factors or influences, the most significant of which is:

A) reactive effects of the experimental procedures
B) contemporary history
C) pretesting procedures
D) statistical regression

Answer 2

statistical regression

Question 3

Referring to the study in item 2 again, for what additional reason might the conclusion be unjustified?

A) There was too small a number of experimental variables being studied to permit an adequate analysis of relationships.
B) There was no adjustment for maturational processes.
C) The classes using the new method were all from one school which would allow the operation of extraneous variables.
D) There was no comparison group.

Answer 3

There was no adjustment for maturational processes.

Question 4

Which of the following is a question of internal validity in experimental research?

A) How widely can the results of the experiment be generalized?
B) Was the sample representative of the population?
C) Has maturation influenced the dependent variable?
D) Did the instruments permit valid interpretations?

Question 5

Subjects performing well merely because they are being observed (and not necessarily because of any effect of treatment) are considered to be under the influence of:

A) the Hawthorne effect
B) the novelty effect
C) the halo effect
D) none of the above

Question 6

Internal validity often receives little attention in research proposals, perhaps because:

A) it is not very important
B) there is no agreement as to what it means
C) there is no way to handle problems of internal validity
D) a study can be carried out without considering it

Question 7

The threat of "subject characteristics refers to differences among subjects on:

A) a dependent variable
B) an independent variable
C) an extraneous variable
D) a treatment variable

Question 8

A threat to internal validity is the same as:

A) bias on the part of the researcher
B) an alternative explanation of results
C) a limitation on the generalizability of results
D) poor internal consistency of scores

Question 9

Loss of subjects is a threat to internal validity because:

A) it reduces the size of the sample
B) those "lost" may differ on the dependent variable from the remainder
C) it reduces the power of statistical tests of significance
D) it means that results of the study are less generalizable

Question 10

Which of the following is not an example of a possible location threat?

A) Experimental and control groups are in different schools.
B) Experimental classes are smaller than control group classes.
C) All tests are given in a noisy gymnasium.
D) Some parents are interviewed at home, others at school.

Question 11

Instrument decay refers to the possibility that:

A) test content becomes "dated" during the course of a study
B) data collectors are lost during a study
C) records of data are lost during a study
D) the instrument is systematically altered during the course of a study

Question 12

Loss of subjects is a serious threat to internal validity in an experimental study if:

A) the sample was too small to begin with
B) the loss is the same for all groups
C) more are lost from one group
D) losses are due to dislike of the researcher

Question 13

Data collector characteristics may be a threat to internal validity if:

A) they are related to measurement of the dependent variable
B) they are related to the independent variable
C) more than one person collects the data in a study
D) all of the above

Question 14

Which is the least-effective means of controlling data collector bias?

A) Extensive training of the data collectors
B) Using norm-referenced instruments
C) Preventing data collectors from knowing the purpose or hypothesis of the study
D) Using a large sample

Question 15

Loss of subjects in a study can best be handled by:

A) random replacement of subjects who are lost
B) providing demographic data on those who are "lost"
C) making statistical adjustments to the data
D) making every effort not to lose subjects

Question 16

Which of the following might be a possible testing threat to the internal validity of a study?

A) A data collector allows some students additional time on a test.
B) Students are tested before and after a film on prejudice.
C) Experimental and control groups are given different instruments.
D) Some students mark answers on the test booklets.

Question 17

A history threat to internal validity refers to the possibility that:

A) the dependent variable is affected by past experiences of the subjects
B) historical events are incorrectly reported
C) unknown events occurring during the study may affect the dependent variable
D) the prior history of the researcher may lead to unconscious bias

Question 18

An implementer threat to internal validity in a study comparing two methods is best controlled by:

A) allowing teachers to select the method they will use
B) insisting that each teacher use the assigned method
C) making sure that all teachers are equal in ability
D) using many teachers with each method

Question 19

When the same questionnaire is scored for opinions on two or more topics, a threat to internal validity may exist because:

A) respondents may psych out the researcher's hypothesis
B) the questionnaire becomes too long
C) some items may be similar to other items
D) reliability of both scores is likely to be reduced

Question 20

In an experimental study, a subject characteristics threat may be present if:

A) experimental and control groups differ in size
B) age is related to the dependent variable
C) age is the independent variable
D) both A and B are correct

Question 21

In which of the following is a history threat least likely to occur?

A) Experimental and control groups are in the same classroom.
B) Experimental and control groups are in different classes in the same school.
C) Experimental and control groups are in different schools.
D) Experimental and control groups are limited only to females.

Question 22

With which of the following findings would a maturation threat be of most concern?

A) "Skills in long division showed marked improvement after three weeks of instruction in computer use."
B) "Third-grade reading scores demonstrated substantial gains after six months of perceptual training."
C) "Knowledge of history improved greatly after a year of inquiry training."
D) "Foreign language usage was greatly enhanced by two years' study in a language laboratory."

Question 23

Which of the following illustrates an attitudinal threat to internal validity?

A) Teachers resented having to try a new curriculum approach.
B) The new curriculum was too difficult for the grade level in which it was used.
C) Students resented not receiving the new curriculum.
D) Test questions did not validly assess the new curriculum.

Question 24

A regression threat to internal validity refers to the fact that:

A) groups selected at extremes will score closer to the average on retesting
B) students often forget what they learned during an intervention
C) a relationship is likely to be smaller when a study is replicated
D) data collectors can influence scores by their behavior

Question 25

Which threat would not apply to a correlational study?

A) Subject characteristics
B) Data collector bias
C) History
D) Testing

Question 26

In qualitative research, internal validity:

A) is irrelevant
B) has a completely different meaning
C) is the same as instrument validity
D) takes a different form than in quantitative research

Question 27

Matching

-Instrumentation threat

A) Age, gender, and class
B) Unreturned surveys
C) Classrooms with fewer resources
D) A fatigued researcher scoring data
E) An earthquake

Question 28

Matching

-Mortality threat

A) Age, gender, and class
B) Unreturned surveys
C) Classrooms with fewer resources
D) A fatigued researcher scoring data
E) An earthquake

Question 29

Matching

-Subject characteristics threats

A) Age, gender, and class
B) Unreturned surveys
C) Classrooms with fewer resources
D) A fatigued researcher scoring data
E) An earthquake

Question 30

Matching

-History threat

A) Age, gender, and class
B) Unreturned surveys
C) Classrooms with fewer resources
D) A fatigued researcher scoring data
E) An earthquake

Question 31

Matching

-Location threat

A) Age, gender, and class
B) Unreturned surveys
C) Classrooms with fewer resources
D) A fatigued researcher scoring data
E) An earthquake

Question 32

The Hawthorne effect deals with subjects who want to intentionally undermine the purpose of a study because they are suspicious of the researchers.

Question 33

Studies in special education are particularly vulnerable to regression threats.

Question 34

To avoid a possible implementation threat, a researcher should not be one of the individuals who implement a method in an intervention study.

Question 35

A history threat is one that a researcher cannot anticipate.

Question 36

Controlling threats to the internal validity of a study is best done once the data have been collected.