Buy essay on True experimental designs

Question 2. How can researchers ensure that there are no biases in the assignment of participants to groups in an experiment?
Researchers use the method of randomization as a means of avoiding bias; pretest-posttest randomized control group design for eliminating biases during dividing participants into groups is used – half of the participants are randomly selected to experimental group, and the remaining half of the participants are control group. Such random design is referred to as true experimental design.
Question 7. What is a potential drawback to the Solomon randomized four-group design?
For Solomon four-group design it is necessary to have quite a large pool of respondents in order to achieve reliable results; Solomon’s division should not be used when the number of respondents is less than 48; it is recommended to have many participants to achieve high reliability in this method.
Topic 38. Threats to internal validity
Question 2. Suppose an experimental group is being taught letters of the alphabet as a treatment. At about the same time, the students are watching an educational program on television, from which they learn the names of the letters. What is the name of the threat this program illustrates?
Such situation represents a threat to internal validity which is called history – some other environmental influence besides the treatment (in our case, educational program) that has an influence on the participants between the pretest and posttest.
Question 3. If observers are more tired and less astute when making posttest observations than when making pretest observations, what threat is operating?
Situation when observers are in a different state (e.g. more tired and less astute) during posttest than during pretest represents a threat to internal validity depending on instrumentation; in this case, the instrumentation during pretest and posttest is different, and thus the results of the experiment cannot be regarded as valid.
Question 4. What is the name of the threat posed by nonrandom assignment of participants to experimental and control groups?
The threat of using intact groups (when the participants are not assigned to groups randomly and/or existing groups are used) is called selection threat.
Question 6. Under what circumstance will statistical regression operate?
The threat of statistical regression appears when the participants are selected on the basis of their scores according to some testing, and the “extreme” part of them is chosen for treatment. In this case statistical regression may lead to a validity threat.
Topic 39. Threats to external validity
Question 3. What is the name of the threat that warns researchers to be careful in generalizing the results to a population when an experiment is conducted on a nonrandom sample?
If a nonrandom sample is used for experiment, it means that there exist selection bias; “selection bias” is also the name of the threat to external validity which does not allow to generalize the results to the whole population if the sample was nonrandom. The way to avoid this threat is to use randomized sample which is unbiased by definition.
Question 6. If a pretest causes a change in participants’ sensitivity to a treatment, what threat is operating?
The threat when pretest changes the sensitivity of the participants to the treatment is called pretest sensitization, or reactive effect of testing. This threat is usually eliminated by using experimental design without pretest.
Topic 40. Preexperimental designs.
Question 1. Are preexperimental designs valuable for identifying cause-and-effect relationships?
No, preexperimental designs are not valuable for analyzing causes and effects, because they give the researcher no information for comparison between groups or between the pretest and posttest condition. These designs might be used for pilot studies but should not be used for planning true research experiments.
Question 4. If a researcher gives a pretest on knowledge of a child abuse to a group of social workers, then gives them a series of seminars on child abuse followed by a posttest, what is the name of the preexperimental design the researcher used?
In this case, the researcher uses the preexperimental design called one group pretest-posttest design (O-X-O).
Topic 41. Quasi-experimental design.
Question 4. What is a major advantage of the equivalent time-samples design?
The major advantage of the equivalent time-samples design is the fact that the same group of participants is regarded both as control and experimental group; thus, the researchers may be sure that control and experimental group have the same characteristics.
Question 5. What is a major disadvantage of the equivalent time-samples design?
The major disadvantage of the equivalent time-samples design is that it is highly vulnerable to one external validity threat – multiple-treatment interference.
Question 6. In psychology, what is ABAB design?
ABAB design in psychology is the other name for equivalent time-samples design when alternating treatments are used one after another; it indicates that one treatment (A) is given after another (B), and vise versa.
Topic 42. Confounding in experiments.
Question 2. What is the name of the effect that refers to the possibility that the control group might become aware of its “inferior” status and respond by trying to outperform the experimental group?
The name of the effect when the control group might try to outperform the experimental group because they are informed of their control group status, is John Henry effect.
Question 3. What is the formal name of what is characterized as the “attention effect” in this topic?
The formal name of the so-called “attention effect” is Hawthorne effect.
Question 6. In what type of experiment do neither the participants nor the individuals dispensing the drug know which is the active drug and which is the placebo?
The type of experiment when none of the participants and none of the individuals dispensing the drug in the experiment know where the active drug is and where placebo is, is called double-blind effect (usually applied in medication treatment experiments).
Question 7. A “demand characteristic” is a cue that lets participants know what?
“Demand characteristic” takes place when there is a cue letting the participants know the outcomes that the researchers will expect from them; in this situation, confounding appears because the participants might try to “please” the researchers and answer in order to match the expected outcomes.
Part 2.
1. Identify the type of design that you have chosen for your study and discuss the nature of the design as it relates to your study.
Among the true experimental research designs there were two giving best reliability and bias elimination: Solomon randomized four-group design, and pretest-posttest randomized control group design. For the research, I have chosen the classical design used for exploring cause-and-effect relationships – pretest-posttest randomized control group design. This research design allows to avoid biases in the experiment. Since the number of participants (500) is enough to maintain reliability for both groups, such research design is the most reasonable. For the depth of the analysis, I have designed quantitative pretest and posttest measurements (Harkness classroom engagement and MAT scores) as well as qualitative survey (School Sleep habits Survey). Most evaluation will be done on the basis of quantitative results, since the study is of quantitative origin. However, qualitative survey will help to understand the changes related to later school start time more deeply, and perhaps provide a basis for further research. I did not choose Solomon randomized four group design because of two reasons: firstly, the size of the experimental group will be reduced from 250 to 125 participants, which will reduce reliability, and secondly, pretest sensitization in the experiment is expected to be low because the measurements will be done in a year (and thus, one of the main advantages of Solomon design over pretest-posttest randomized control group design loses its value).
2. Identify all threats to internal validity.
First of all, it should be mentioned that proper research design can eliminate most threats to internal validity. In our case, such threats as history, instrumentation, testing and selection are eliminated. Indeed, since one year will pass between the tests, the possibility of historical interference of them is very low; instrumentation is already known and will not be changed; the threat called testing is also not applicable due to the 1 year time period between the pretest and posttest. The threat of selection does not appear because the participants for control and experimental groups are chosen randomly. Since the participants are selected randomly from the whole school, the threat of statistical regression is also not applicable. There are two threats which can be slightly applicable to our case: maturation and mortality. During 1 year the students will naturally mature; however, MAT tests and Harkness classroom engagement scores take this maturation process into account. School Sleep Habits Survey should also be corrected in order to divide changes caused by the experiment from maturation consequences; however, since the qualitative survey will be used as supplementary material, the threat of maturation is also of low value concerning the experiment.
The threat of mortality (students leaving or entering the school during the experimental year) can indeed influence the experiment; however, it is difficult to predict whether there will be a significant number of such students. This threat should be analyzed after the posttest; perhaps, the students that did not stay in the school longer than 2 years should not be included in the experiment.
3. Identify all threats to external validity.
Possible threats to external validity include: selection bias, reactive effects of experimental arrangements, pretest sensitization and multiple-treatment interference. Selection bias is a threat that to a certain extent relates to our experiment: since there was only one school chosen for the experiment, there exists a possibility that the selection sample is biased, and generalizations should be made with caution. In the ideal variant, this experiment has to be repeated in different schools and in different geographic areas in order to provide the possibility is generalization. However, since this school does not have characteristics that differ it significantly from other private schools in this area or in other states, the results of the experiments can be generalized, with taking into account the limitations and possible bias of the study. Pretest sensitization, as it was discussed concerning internal validity, is very low. Multiple-treatment interference possibility is also quite low since school start time is a subject that does not change often; sleep patterns of the participants are suggested to be quite varied from the very beginning and will not be the source of multiple-treatment interference threat.
There is a possibility of reactive effect of experimental arrangements – in our case, it is the possibility that the students whose school start time is changed (they will certainly have to be rearranged in the groups according to random selection) will realize that they are the experimental group and change their learning efforts. Thus, it is necessary to maximally integrate the school start time change into the learning process and provide a reasonable explanation of the changes so that the students would not know about the experiment.