Instructions

For each of the following scenarios determine which of the following four methods is best illustrated: correlational, ex post facto, quasi-experimental, true experimental.

1. A researcher hypothesizes that directive teachers will be more effective than non-directive teachers in motivating students to learn. To test this hypothesis, the researcher administers to a sample of teachers an instrument designed to measure directiveness. Results obtained from this instrument will simply allow for the classification of teachers as either directive or non-directive. After collecting this information, the researcher next administered an instrument designed to measure motivation to the students of those teachers sampled. After obtaining and analyzing this data, the researcher found that students of non-directive teachers were just as motivated as students of directive teachers.

2. A researcher wishes to investigate the effects of cooperative group testing (CGT) on achievement among college students. The researcher decides to use his two educational research classes for the study. The researcher randomly assigns CGT to the Tuesday night class and traditional instruction to the Thursday night class. At the end of the quarter, the researcher found that average achievement scores for both classes were about equal.

3. A researcher wishes to know whether dynamic assessment, static assessment, or a combination of the two (static and dynamic assessment) will increase teachers’ expectations for mentally handicapped children. The researcher randomly assigns the teachers to one of three groups. The first group of teachers views a child via a dynamic assessment procedure. The second group views the same child by a static assessment procedure, and the third group views the child under both procedures. After this viewing all three groups complete an instrument designed to measure the teachers’ expectations of the child’s likely academic performance.

4. An investigator in interested in the examining the possible relation between sex and creativity. He measures the creativity of a sample of boys and girls and finds that boys are statistically more creative than girls. He concludes that boys are more creative than girls.

5. Some college professors felt that the most recently matriculated students were not as bright as their older peers. To test this supposition, they randomly selected 100 freshmen and 100 seniors. Next, they asked the registrar’s office to provide SAT scores for each of the randomly selected students. After comparing the results, they found that seniors averaged about 75 points better than the freshmen.

6. A random sample of 1,000 people was selected and divided into groups of smokers and non-smokers. Medical doctors examined and rated the health of each individual. The ratings given by the doctors ranged from 1 (excellent health) to 5 (extremely poor health). The doctors found that the average health of non-smokers was 3.1 and the average health for smokers was 3.4. The doctors concluded...

7. Following up on the study presented above in 6, the doctors focused exclusively on the smokers for their next investigation. In this study the smokers were classified by the extent of their smoking habit. The classifications ranged from 1 (one cigarette per day or less) to 5 (more than 3 packs per day). Using the health classification scale discussed in 6, the doctors wanted to know whether an association existed between the intake of cigarettes per day and one’s health.

8. A researcher develops a new measure of reasoning ability. Her new instrument rates reasoning ability on a scale that ranges from 1 to 100. She randomly chooses 50 students to participate in her study. She administers the reasoning ability instrument to the students and records their scores. Next, she collects each student’s quantitative GRE and quantitative SAT scores. She then administers the Group Embedded Figures Test (GEFT) to the students and collects each student’s score on this instrument. After collecting and analyzing all the scores, she finds that scores obtained from her reasoning ability instrument are highly related to GRE, SAT, and GEFT scores. These results led her to believe that she has established construct, and to some extent concurrent, validity for her new test.

1. A researcher hypothesizes that directive teachers will be more effective than non-directive teachers in motivating students to learn. To test this hypothesis, the researcher administers to a sample of teachers an instrument designed to measure directiveness. Results obtained from this instrument will simply allow for the classification of teachers as either directive or non-directive. After collecting this information, the researcher next administered an instrument designed to measure motivation to the students of those teachers sampled. After obtaining and analyzing this data, the researcher found that students of non-directive teachers were just as motivated as students of directive teachers.

Ex post facto: Groups were compared (directive vs. non-directive), and the groups were not formed by the researcher or manipulated by the researcher.

2. A researcher wishes to investigate the effects of cooperative group testing (CGT) on achievement among college students. The researcher decides to use his two EDR 751 classes for the study. The researcher randomly assigns CGT to the Tuesday night class and traditional instruction to the Thursday night class. At the end of the quarter, the researcher found that average achievement scores for both classes were about equal.

Quasi-Experimental: The groups were not randomly formed by the researcher, but there was manipulation--the researcher determined which class would receive which treatment.

3. A researcher wishes to know whether dynamic assessment, static assessment, or a combination of the two (static and dynamic assessment) will increase teachers’ expectations for mentally handicapped children. The researcher randomly assigns the teachers to one of three groups. The first group of teachers views a child via a dynamic assessment procedure. The second group views the same child by a static assessment procedure, and the third group views the child under both procedures. After this viewing all three groups complete an instrument designed to measure the teachers’ expectations of the child’s likely academic performance.

True Experimental: The researcher randomly formed the groups and manipulated which treatment each group would receive.

4. An investigator in interested in the examining the possible relation between sex and creativity. He measures the creativity of a sample of boys and girls and finds that boys are statistically more creative than girls. He concludes that boys are more creative than girls.

Ex post facto: Comparison among groups (boys vs. girls), but the groups were not manipulated

5. Some college professors felt that the most recently matriculated students were not as bright as their older peers. To test this supposition, they randomly selected 100 freshmen and 100 seniors. Next, they asked the registrar’s office to provide SAT scores for each of the randomly selected students. After comparing the results, they found that seniors averaged about 75 points better than the freshmen.

Ex post facto: Groups were compared (freshmen vs. seniors), but the groups where not randomly formed by the researcher or manipulated. In this case one may be confused by the random sampling. This is a selection procedure that has no bearing on the assignment of students to groups, that is, the students are either freshmen or seniors, the researcher cannot manipulate this.

6. A random sample of 1,000 people was selected and divided into groups of smokers and non-smokers. Medical doctors examined and rated the health of each individual. The ratings given by the doctors ranged from 1 (excellent health) to 5 (extremely poor health). The doctors found that the average health of non-smokers was 3.1 and the average health for smokers was 3.4. The doctors concluded...

Ex post facto: As with the example above, the participants were randomly selected, but does not mean the groups were randomly formed. In this case, the people selected indicated whether they smoked, so this variable was not manipulated. Groups were compared (smokers vs. non-smokers), but no manipulation existed.

7. Following up on the study presented above in 6, the doctors focused exclusively on the smokers for their next investigation. In this study the smokers were classified by the extent of their smoking habit. The classifications ranged from 1 (one cigarette per day or less) to 5 (more than 3 packs per day). Using the health classification scale discussed in 6, the doctors wanted to know whether an association existed between the intake of cigarettes per day and one’s health.

Correlational: Examination of the relationship between two quantitative variables (health rating and extent of smoking rating). No groups were compared and neither variable was manipulated by the researcher.

8. A researcher develops a new measure of reasoning ability. Her new instrument rates reasoning ability on a scale that ranges from 1 to 100. She randomly chooses 50 students to participate in her study. She administers the reasoning ability instrument to the students and records their scores. Next, she collects each student’s quantitative GRE and quantitative SAT scores. She then administers the Group Embedded Figures Test (GEFT) to the students and collects each student’s score on this instrument. After collecting and analyzing all the scores, she finds that scores obtained from her reasoning ability instrument are highly related to GRE, SAT, and GEFT scores. These results led her to believe that she has established construct, and to some extent concurrent, validity for her new test.