Provide a 4-6 sentence response to the questions below. Give evidence and/or examples to support your answers.
1) Explain the differences between a between-subjects and within-subjects design.
2) Compare and contrast internal validity and external validity.
3) What is a confound and how is it related to internal validity?
4) Differentiate parametric and nonparametric statistics.
5) What is the relationship between external validity and the college sophomore problem?
6) Overview the principle of falsifiability. Why is this concept important to science?
7) Assume you are making a directional hypothesis, are using an alpha level of .01, and have 12 degrees of freedom. What t-critical value should you use to decide whether to retain or reject the null hypothesis? Explain how you arrived at your answer.
8) What are the pros and cons of using an experiment for psychological research?
9) What is counterbalancing? When might a researcher consider using this procedure?
10) Define degrees of freedom. Why is degrees of freedom an important concept in research design and statistics?
1) While between-groups design is the experimental design where the individuals from experimental group that needs to be observed are not the same with the individuals from control group that measure the outcomes of experimental group, the within-subjects design is more focus on individuals than groups and the same individuals pass through all the stages of the experiment, taking iterative measures.
2) The internal validity measures the level of cause-effect relationship through a logic process while the external validity measures the degree of generalization of the results of a study.
3) Confounders are causally factors that can alter the internal validity that measure the level of cause-effect relationship, abnormally increasing or decreasing the degree of association between the cause and effect and leading to incorrect conclusions.
For example, if the study focuses on how drinking coffee can influence the pancreatic cancer (drinking coffee - cause and pancreatic cancer - effect), the confounder that can modify the conclusion of this study is the smoking habit that is associated with the habit of drinking coffee.
4) Parametric and non-parametric statistics use different tests to discover population differences. While parametric tests, such that two sample t-test, can discover population differences only if the differences between the compared samples are normally distributed, the non-parametric tests, such that Wilcoxon Mann-Withney test, can ignore the assumption that the differences between the compared samples are normally distributed.
9) Since counterbalancing is the method that can exclude the threaten of confounders over the internal validity of an experiment, it should be considered by researchers. The disadvantage of this method is that it becomes complicated when the experiments need to consider more than 3 conditions.
10) The variables that vary freely are called,in statistic studies, as degrees of freedom. For example, if you have 4 pens and you want to write with each pen once, in a month, you will have 3 degrees of freedom because 3 weeks you can select what pen you wish but in the fourth week, there will be one pen left to take, hence, you can no more make a selection, but to use the last pen you have. The degrees of freedom can be used when the sample size is not large enough to collect significant observations.