Questions:
1. Define the term quasi-experiment and discuss the pros and cons of this research:
Answer:
A quasi-experimental design is a research methodology that possesses some, but not all, of the defining characteristics of a true experiment. In most cases, such designs examine the impact of one or more independent variables on dependent variables, but without assigning participants to conditions randomly or maintaining strict control over features of the experimental situation that could influence participants’ responses.
Quasi-experiments are most useful when conducting research in settings where random assignment is not possible because of ethical considerations or constraining situational factors. In consequence, such designs are more prevalent in studies conducted in natural settings, thereby increasing the real-world applicability of the findings. Such studies are not, however, true experiments, and thus the lack of control over assignment of participants to conditions renders causal conclusions suspect.
References
Cook, T. D., & Campbell, D. T. (1979). Quasi-experimental: Design and analysis issues for field settings. Boston: Houghton Mifflin.
Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental designs for generalized causal inference. Boston: Houghton Mifflin.
2. Explain the pros and cons of longitudinal, cross-sectional, and sequential designs
Answer:
And because almost every science has someone who cleverly combines things, we have asequentialdesign, also sometimes referred to as a cross-sequentialdesign, which is defined as a combination of longitudinal and cross-sectional designs, by following several differently aged cohorts over time.
Alongitudinal study, like across-sectionalone, is observational. So, once again, researchers do not interfere with their subjects. However, in alongitudinal study, researchers conduct several observations of the same subjects over a period of time, sometimes lasting many years.
In across-sectional study, data are collected on the wholestudypopulation at a single point in time to examine the relationship between disease (or other health related state) and other variables of interest.
3. A researcher has studied subjects' ability to learn to translate words into Morse code. He has experimented with two treatment conditions: in one condition the subjects are given massed practice they spend eight full hours on the task. In the other condition, subjects are given distributed practice; they also spend eight hours, but their practice is spread over four days, practicing two hours at a time. After the practice, all subjects are given a test message to encode; the dependent variable is the number of errors made. The researcher has matched the subjects on intelligence. The results are in the following table. Decide which statistical test would be appropriate, carry out the test, and evaluate the outcome. Assume a significance level of .05 and that the direction of the outcome has not been predicted
Mass Practice Distributed Practice
S1 6 S1 5
S2 4 S2 3
S3 3 S3 2
S4 5 S4 2
S5 2 S5 3
Answer:
Non-normally distributed variables—non-parametrictests: If the parameter of interest is not normally distributed, but at least ordinally scaled,non-parametricstatistical tests are used. One of these tests (the “rank test”) is not directly based on the observed values, but on the resulting rank numbers.
4. Chuck w**** is very excited about the with-in subjects approach. "Now I'll never need to run large numbers of subjects again. "He says, however, Chuck has forgotten that with-in subjects designs maybe a)useless b)impossible c)confounded by order effects, or d) impractical when excessive subject time spent in an experiment makes data inaccurate. Give an example of each of these four objectives
5.Describe a two-matched groups design. How is the matching done?
Answer:
One important type of experimentaldesignis amatched-subjectsdesign, also called amatched-group design, which is when subjects arematched on some variable that might be affecting the dependent variable and then split intotwoor more groups.
Matched group design (also known as matched subjects design) is used in experimental research in order for different experimental conditions to be observed while being able to control for individual difference by matching similar subjects or groups with each other.
6. Bill Boards "lording" his SAT score over his head Rhonda D*****, who took the ACT." You only got a 25 in math, "he chortled, " while I got a 300 in math." Given that the SAT has a u of 500 and a o of 100, and the ACT has a u of 20 and a o of 5, what is wrong with Bill's logic (give the answer in both z scores and percentile ranks).
Answer:
Standardized tests are biased.
7. Explain how a one-way analysis of variance works. How do you use between and with-in group variability?
Answer:
Theone-way analysis of variance(ANOVA) is used to determine whether there are any statistically significant differences between the means of two or more independent (unrelated) groups (although you tend to only see it used when there are a minimum of three, rather than two groups). Within-groupvariation (sometimes called errorgroupor errorvariance) is a term used in ANOVA tests. It refers to variations caused by differences within individualgroups(or levels). In other words, not all the values within eachgroup(e.g. means) are the same.
8. Explain the value of reversal designs . ( ABA designs) in single-case research.
Answer:
Reversal Design: any experimental design in which the researcher attempts to verify the effects of the independent variable by “reversing” responding to a level obtained in a previous condition; encompasses experimental designs in which the independent variable is withdrawn (A-B-A-B) or reversed in its focus [e.g. DRI/DRA] (Cooper, Heron, & Heward, 2007).
Withdrawal Design: often used as a synonym for A-B-A-B design; also used to describe experiments in which an effective treatment is sequentially or partially withdrawn to promote the maintenance of behavior changes (Cooper, Heron, & Heward, 2007).
9.After watching nursery-school children, Ken G**** wants to test the hypothesis that some toys are more fun to play with than others. He decides to compare "fun" toys (blocks) with "unfun" toys (stuffed animals). He also wishes to see if there is a sex difference, as well, so sex is added as an independent variable. A) what kind of design is needed B) Diagram it out C) Assuming 20 subjects are needed per cell, how many subjects are needed for this study?
Answer:
When planning data collection for a qualitative research study, whether for a Ph.D. dissertation or a new business model, researchers often ask how many participants are enough? The answer is, enough is the amount where additional participants don’t provide any additional insights. We call this phenomenon “saturation.” You reach saturation when you are no longer learning very much (if anything) from each subsequent interview, observation, etc. So, how many do you “typically” need to reach saturation? Good question.
10 For each of the following examples, explain whether the researcher has made a correct decision or has made a Type 1 or Type 2 error. Explain why.
1.) Dr. G rejects the null hypothesis although the independent variable had no effect:
It is Type 1 error because he rejected the null hypothesis.
2.) Dr. Rejects the null hypothesis when it is false.
This is a Type 2 error is when someone fails to reject a false null hypothesis.
3.) Although the independent variable had an effect, Dr. E does not reject the null hypothesis.
This is a Type 2 error because Dr. E did not reject the false null hypothesis.