home / study / math / statistics and probability / statistics and probability qu
ID: 3336612 • Letter: H
Question
home / study / math / statistics and probability / statistics and probability questions and answers / anderson et al (1998) conducted a study to explore the "weapons effect" (i.e., the finding ... Question: Anderson et al (1998) conducted a study to explore the "weapons effect" (i.e., the finding that p... Anderson et al (1998) conducted a study to explore the "weapons effect" (i.e., the finding that people tend to behave more aggressively in the presence of weapons). They take this a bit further by hypothesizing that people are quicker, on average, to say aggressive words after reading weapons names, than after seeing other non-weapon words. 35 undergraduates (19 men and 16 women) at the University of Missouri-Columbia volunteered to participate in this study. Participants were told by the researcher that the study was being conducted to test reading ability for various types of words. The task involved reading a word on a computer screen (the authors called this word the “prime” and it was either a weapon name like “shotgun” or a non-weapon word like “butterfly”), then saying a “target” word (either an aggressive word like “assault” or a nonaggressive word like “joke”) out loud when prompted. Each subject repeated this task 192 times so that all the possible “target” words (i.e., 24 aggressive ones and 72 non-aggressive ones) could be repeated for a weapon and a non-weapon “prime” word. The order in which the target words were presented was randomly determined for each study participant. The times it took for the participants to say the “target” words verbally after they were shown were recorded so the mean times could be compared across the different combinations of “prime” and “target” words. More information on this study is available on pp 309-210 of the paper (although you do not need to read the paper to answer the assignment questions): Anderson, C.A., Benjamin, A.J., & Bartholow, B.D. (1998). Does the gun pull the trigger? Automatic priming effects of weapon pictures and weapon names. Psychological Science, 9, 308-314. Available at http://journals.sagepub.com/doi/abs/10.1111/1467-9280.00061
How and why did these researchers use i. randomization, ii. replication and iii. control in their study?
Explanation / Answer
According to Sir R.A. Fisher the three basic principles of Design of experiment are randomization, replication and local control. To capture the variability due to a factor (or treatment), we need to eliminate other possible causes of variation from the dataset. For that these three principles are used. These are explained according to the given situation below:
Here 35 undergraduates were included as the study participants. Among them 19 were men and 16 were women. The order in which the target words were presented was randomly determined for each study participant. Each subject repeated this task 192 times so that all the possible “target” words (i.e., 24 aggressive ones and 72 non-aggressive ones) could be repeated for a weapon and a non-weapon “prime” word. The random order in which the target words were presented for each study participant, is necessary for randomization. Total 96 words (24 aggressive & 72 non-aggresive) were repeated twice (total 192 words) so that each word were repeated exactly twice. That means the experiment is replicated twice. The replication is necessary because of the necessity to capture the variability due to the experimental process. In each replication, if the ordering of the treatment allocation remains same (here the presentation of target word), there may exist some effect which reduce the variability from one replication to another. Here the randomization is maintained by random ordering of presentation of target words. Last of all, there may exist some variation due to gender effect which is captured here by choosing the experimental units into two groups -- men and women. This is the local control.
Related Questions
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.