Pretest
User ID: dheise Attempt: 1 / 5 Out of: 14
Quiz Started: Jan 06, 2000 16:00 Quiz Finished: Jan 12, 2000 0:12 Time Spent: 128 hrs., 11 min., 4 sec.



Write a response to each of the following items. If you have doubts about the correctness of your answers, share them with other members of your small group or one of the instructors of the course. If many of your responses are not correct, it indicates that you probably need more study of research procedures than will be provided in this course. In this case you should purchase a general research methods textbook (such as Borg & Gall) to study this quarter along with the other readings of the course.


Give an example of a null hypothesis.
No. Student Response:
1. There is no real difference between PCs and Macs.

General Feedback:
Principle: A null hypothesis usually indicates that there is no difference between groups or no relationship between variables. Example: There is no difference in test scores between men and women.

Score: 1 / 1

Give an example of an operational definition.
No. Student Response:
2. Software quality can be measured as the number of known bugs per 1000 lines of code. While this definition is probably adequately clear for others to get reproducible figures, it is not really a useful measure of software quality. Ease of use and applicability to user needs are much more important in assessing software quality, but are also much more difficult to measure.

General Feedback:
Principle: An operational definition indicates how you intend to measure the variable. Example: Depression is the number of times you cry per week.

Score: 1 / 1

What is ERIC used for?
No. Student Response:
3. ERIC (Education Resources Information Center) is a large repository of education-related literature. It is useful to education practitioners and researchers alike to learn about research findings in education, to guide and inform in the development of policies and practices as well as leading to new research.

General Feedback:
ERIC is a computerized data base used to search for educational literature.

Score: 1 / 1

Give an example of a stratified random sample.
No. Student Response:
4. I used WebCT to give quizzes in a Management Information Systems course. I had a test bank of 35 True and False questions and 35 Multiple Choice questions for each chapter. Each quiz was to have 20 questions chosen randomly, and students could take each quiz up to 5 times, but with new randomized question sets each time. I wanted to be sure there were 10 True and False and 10 Multiple Choice questions each time, so I divided the 70 possible questions for each chapter according to question type, and randomly sampled 10 from each type to make up the 20 questions.

General Feedback:
Principle: A stratified random sample is a sample in which the elements are randomly selected from many subgroups of the total population. Example: Taking a random sample of 50 men and 50 women.

Score: 1 / 1

Give an example of a purposive sample.
No. Student Response:
5. Rhoda Sommers came and spoke to us during a class in Qualitative Research during summer in 1999. Her dissertation was about characteristics of effective middle school teachers. Her approach was to become knowledgeable in the area by reviewing the literature, then she defined a process for finding effective teachers according the knowledge gained. This is purposive sampling.

General Feedback:
Principle: A purposive sample is a sample in which subjects are chosen specifically for information they can provide. Example: Select the four high school principals who have been the most out-spoken concerning a certain issue, with two being in favor and two against the issue.

Score: 0 / 1

What is the main advantage of longitudinal research?
No. Student Response:
6. This kind of research allows data to be collected from the same population at more than one point in time. For instance, one approach to trying to determine the effects of smoking would be to take a sample spanning ages from young to old, and to try to draw conclusions from the proportions of smokers in each age group. But a better approach would be to select a sample, consisting of smokers and non-smokers, and to monitor them throughout their natural lifespan. One thing this does is to reduce the number of uncontrolled variables.

General Feedback:
The advantage of longitudinal research is that since you are studying the same subjects for a long period of time you can study growth in the subjects.

Score: 1 / 1

What is the purpose of having a control group in a research study?
No. Student Response:
7. If you are trying to demonstrate a cause and effect relationship, you have to be able to show that when everything else is the same except the cause you are studying, the effect occurs only when the cause occurs. A control group is the same as the experimental group in all respects except the factor(s) being experimented with.

General Feedback:
A control group is used so you can evaluate whether changes in the experimental group(s) are due to the treatment which the control group does not receive or to other factors which are similar for the control group and the experimental group(s).

Score: 1 / 1

Give an example of triangulation as used in qualitative research.
No. Student Response:
8. Data being gathered via qualitative research methods is subjective, and is easily influenced by numerous factors that are hard to control. Triangulation is used to try to measure the same factor in different ways to reduce any bias there might be in the measurement. A classic approach in qualitative research is to collect data via both interviews and questionnaires.

General Feedback:
Principle: Triangulation is gather the same data using many sources or using many methods to establish the reliability and validity of the data gathered. Example: Asking the student, the student's teacher and other students in the student's class to evaluate how a student participates in class.

Score: 1 / 1

What are the advantages and disadvantages of participant observation?
No. Student Response:
9 Participant observation, such as living among gorillas being studied, allows a richness of information to be discovered that may never be learned any other way. However, the closer the association between the observer and the observed, the greater the risk that the behavior being observed is being influenced by the observerís presence.

General Feedback:
By becoming part of the group that you are observing, you can participate in the dynamics of what the group is experiencing from within the group, observing things that an "outsider" would never be aware of.

Score: 1 / 1

Why is informed consent important in research?
No. Student Response:
10. Privacy, confidentiality and ethics all have to be considered when dealing with human research subjects. Sometimes, the information being used may be publicly and readily available. But whenever there is any doubt, formal informed consent should be obtained to protect the rights of the individual. Research proposals should always be reviewed to ensure that this is done.

General Feedback:
Human subjects should not be subjected to research conditions that might have a detrimental effect on them unless they (or their guardians) have given their consent, after being informed of what will happen to them and the potential consequences.

Score: 1 / 1

How can a test be made more reliable?
No. Student Response:
11. By appropriate application of statistical principles: in choosing the sample size, to ensure it is not too small; in the methods used for selecting the sample, to ensure that it represents the population with respect to attributes being studied; in the methods used for designing the study, and assigning members of the sample to groups within the study (eg control group)

General Feedback:
The most common ways of increasing reliabiliy of measurement are to have more samples of the trait being measured (longer tests or more observations) or to objectify or standardize the measurement procedure and/or scoring (e.g., check list, multiple choice items, video tape observations).

Score: 0 / 1

Give an example of evidence that would support the construct validity of a test instrument.
No. Student Response:
12. statistical soundness in the design, application of statistical principles; observations support the predictions

General Feedback:
Principle: Construct validity is measured by determining whether the results of your measurement are consistent with other data that are theorized as being related to what is being measured. Example: An IQ test that gives verbal and quantitative scores should expect the verbal scores to be more predictive of verbal-type schoolwork and the quantitative score to be more predictive of quantitative-type schoolwork.

Score: 1 / 1

What might cause a research study to lack internal validity?
No. Student Response:
13. Attempts to measure the same thing via triangulation could miss the mark, and not measure the same thing at all. This could be because the test instruments were poorly designed, or errors were made in the sampling, or other reasons.

General Feedback:
Principle: Studyies lack internal validity when the conclusions are not correct because of other factors not studied in the research. Example: If a study claimed that men get higher grades than women but the sample was not representative because it had more "bright" men than "bright" women than in the population, then IQ is the cause, not gender and the study lacks internal validity.

Score: 1 / 1

What might cause a research study to lack external validity?
No. Student Response:
14. The factors being measured could all be influenced in a significant way by variables not being measure or controlled. This could be through lack of proper controls, or inadequate sample size or selection. If the study were too small, or the sample selection too narrow, it would be improper to make broad generalizations.

General Feedback:
Principle: A study lacks external validity if the sample used in the study is not similar to the group to which the conclusions are made by the authors of the study or by others. Example: A study based on children in the USA may not have external validity when applied to children in France.

Score: 1 / 1