Research with People by Nigel Holt Ian Walker

Author:Nigel Holt, Ian Walker [Nigel Holt, Ian Walker]
Language: eng
Format: epub
Tags: Social Science, Methodology, Research
ISBN: 9781350312494
Google: r_tGEAAAQBAJ
Publisher: Bloomsbury Publishing
Published: 2009-03-31T02:37:33+00:00

Nothing wrong there. I’m happy with that

We’re not happy with it and neither should you be.5 We’ve summarized some of the issues on page 126, but it’s worth going over them in a little more detail here. First, the word-search puzzle. It is not clear from this whether everyone received the same one. If they did not, then a confound may have been introduced. If different groups are given a different word-search puzzle then any differences in performance between groups cannot be attributed entirely to the type of management feedback the participants were being given.

Second, everyone was in the same room, and everyone was, therefore, aware of what was going on around them. If those receiving negative criticism could hear that others on a table nearby were receiving positive encouragement they may have responded in a way not anticipated or controlled for by the researchers. It is possible their self-esteem may have suffered with the belief others were doing better than they were: they might have tried even harder if they knew others could hear the comments they were receiving. Anything like this, that is not controlled for but which may influence the independent variable, is a possible confound and must be carefully designed out of the procedure. The way to do this, of course, is to test each individual separately, in the same room, with the same surroundings. This way they would be unaware of the different groups and how different people were receiving different comments from the researchers, or indeed that some people were not subjected to any comments at all.

This brings us to another issue with the design as indicated in our ‘think-through’. Twelve different researchers were used. Presumably each was carefully trained in the procedure, and how to go about providing the type of comments they were giving, but how can we be absolutely sure their comments were consistent? One researcher may have been harsher than another in giving negative criticism, or, similarly, one may have been much nicer than another in giving positive encouragement. We are also told that the ‘official-looking’ researchers are ‘men and women’. This may also be an issue. Men may be perceived as more aggressive than women and so this may be a factor in the effect their comments may have. The best thing to do would be to use the same researcher each time, thereby keeping ‘presence’ and ‘charisma’ and ‘natural authority’ exactly the same for each participant.6

How about the number of people in each condition? In our example we had six people in each level of our independent variable. Are six really enough? The major problem with independent-samples designs is that individual differences might affect the dependent variable. How have the researchers controlled for the possibility of the world word-search puzzle champion finding their way into the ML2 group? Their world-class word-search puzzling would make the ML2 scores look very good indeed. When you design the procedure you need to do your best to control for such issues, and we’ve described how you might go about doing this elsewhere in the book.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.