Mini-Report: research methods - Unconscious Attitudes study

Method

Participants

One hundred and twenty stage 1 psychology undergraduates at the University of Plymouth voluntarily participated in the study.

Materials

A computer was used to present the Implicit Association Test during the experiment.

Design and Procedure

Participants completed five sections of trials measuring their associations with elderly people. Before each section participants were given instructions on how to proceed. Although unaware of this, participants completed the sections in two different orders. Half completed the sections in the order 1,2,3,4,5 while the other half completed them in the order 1,4,5,2,3. This was to counter balance and rule out the chance of order effects occurring. The time taken to complete each section was recorded. The first section of trials showed a series of faces that participants identified as young or old using the allocated keys on the computer keyboard. The second and fourth sections asked participants to identify positive and negative words in the same way. The fifth and third sections used alternating faces and words. In the third section participants responded to the old faces and negative words with the same key and young faces and positive words with the same key, then in section five they responded to old faces and positive words with the same key and young faces and negative words with the same key. These two trials were the most important for comparison in measuring associations with the elderly.

Results

The results for this experiment were obtained through the IAT computer program, which measured the time in milliseconds that participants took to complete section five (where old faces and positive words were placed together) and section three (where old faces and negative words were placed together). The computer also calculated the difference between the results for these two sections for each participant. If a participant was faster in completing section three, this could imply they have unconscious negative views toward the elderly.

The descriptive statistics for these two sections and the difference between them are illustrated in Table 1.

Table 1: Mean and standard deviation for sections three and five and the difference between the two.

Section | M | S.D.

Section 3 | 800.21 | 166.20

Section 5 | 981.74 | 208.70

Difference | -181.53 | 161.40

These results show that on average participants preformed faster in section three than in section five as predicted. As participants were faster when the negative words were paired with old faces this implies unconscious negative attitudes toward the elderly. The standard deviation also shows that in section three the results were closer for all participants than in section five where they were more varied.

Discussion

On average, participants responded faster when associating old and negative concepts than old and positive ones. On average, participants found it easier to relate old age with negative concepts. The results suggest that participants had stronger implicit associations between old age and negativity, and this means they were prejudiced towards old age rather than youth.

Participants' responses for old and negative associations had a smaller standard deviation than the inconsistent block. This suggests that the strengths of participants' implicit association were far more consistent towards old age and negativity than old age and positivity.

These results could be further tested for statistical significance and further investigation could be used to determine if there is a link between explicit behaviour and implicit associations, as in the McConnell study.

Whether or not the IAT shows actual prejudice is based on the acceptance of leaping assumptions. The faster someone reacts when relating age and negativity may not show that they are prejudiced. They may just have a stronger schema as they have had more contact with negative old concepts from the world around them, i.e. the media.

Participants took part in the experiment in the university's psychology study rooms. These rooms are not silent working areas so noise from other students could have distracted them. This could have slowed reaction times due to participants' having dual-attention when taking part.

If the IAT is a good predictor of prejudice behaviour it could be extremely useful in HR assessments to determine the suitability of candidates for jobs, especially in the public service sector. Applicants' likelihood to show prejudice behaviour could be measured, whether it be towards age, race or sex.

General

This section is well structured, and includes all the significant elements of the method. To enable replication of the trials in another context, a full report might contain more detail.

View linked text

Quality: Structure

The structure of this methods section is good and clearly signposted. In this type of report the methods section has three sub-sections: "Participants", "Materials" and "Design and Procedure".

View linked text

All tables (and figures) should be referred to in the text at some point. Tables and Figures are numbered separately (e.g. Figure 1, Table 1 etc.)

View linked text

Titles for a table usually go above the table (and titles for a Figure go below). The text "Table 1" is in standard font and the actual title should be in italics. The reverse is true for Figures.

View linked text

This provides a good opening topic sentence that introduces what is to follow

View linked text

Function: Describe

All the relevant details about the participants have been given here extremely concisely, including the population from which they were drawn. Notice that numbers are written out in text where something is being counted (one, two three students"), but given numerically (sections 1,2,3") where something is being labelled. This is good practice and helps readability of texts containing a lot of numerical information.

View linked text

In a full practical report, further information about the IAT tool could be presented. The tool itself or associated materials e.g. the photographs and questions used might be given in an appendix.

View linked text

These are important details to include when describing a study. In a full report, details of the instructions could be summarised or given in an appendix.

View linked text

This sentence describes a refinement of the method and clearly explains the reasons for it being used.

View linked text

It is important in a study of this kind to note that the dependant variable (in this case time) "was recorded".

View linked text

This describes the five sections of the trial process. Because of the complexity of factors involved, in a longer report it would be worth providing some visual or tabular support for this textual description.

View linked text

This final sentence clarifies the significance of the results which will be reported in the following section.

View linked text

This piece of writing very clearly summarises the nature of the data collected. It also briefly reminds the reader of the two important sections in the experiment and what they involved.

View linked text

Here the writer reminds the reader of the hypothesis of the study and clarifies what the expected results would mean.

View linked text

This sentence clearly and concisely summarises what the results show.

View linked text

The Standard Deviation (S.D.) has to be presented alongside all Mean Scores. If there are sizeable S.D. differences between conditions, it is worth drawing the reader"s attention towards these differences as has been done here.

View linked text

A discussion section should always start with a brief summary of the results as has been done here.

View linked text

This sentence describes the differences in standard deviation well, and brings it to the attention of the reader.

View linked text

This now starts to give a more specific indication of the points to be discussed

View linked text

Other

When presenting summary statistics, if the Mean is presented, then it is important to make sure the Standard Deviation is also included. Abbreviating "Mean" to "M and Standard Deviation to S.D., as has been done here, is in accordance with APA presentation guidelines.

View linked text

The numbers here have been rounded to two decimal places. This is in accordance with APA presentation guidelines.

View linked text

This piece of reflection helps to contextualise the topic, as well as giving it greater relevance

View linked text

Function: Analyse

Interpretation of the results should always be left to the discussion section. However, it is acceptable to include a brief statement of whether the results were in the direction predicted or not.

View linked text

This statement goes beyond simply stating the results. Because of this, it should be placed in the discussion section. Interpretation or discussion of what the results mean should not be included within a results section.

View linked text

Here is a brief interpretation of what the results mean.

View linked text

This sentence offers further interpretation and adds weight to the earlier interpretation that the participants had strong implicit associations between old age and negativity.

View linked text

As this is a mini-report no statistical analysis is required, but it is good to demonstrate an awareness of this possibility and to recommend that the results should be examined in this way.

View linked text

This is a good concluding paragraph, giving a summary of the theoretical and real-life implications the results have.

View linked text

This piece of text starts to compare and contrast the two approaches, which is clear for your use of 'however'. As such you are starting to provide some analysis in your argument. What could have strengthened this would have been the inclusion of some more literature, i.e., perhaps about the student to staff ratios of mainstream and SEN schools!

View linked text

Further analysis, helped by the use of an example

View linked text

Function: Evaluate

This is a good evaluative paragraph. The writer has considered other possible explanations beside the one set out in the literature, which could also explain the pattern of results.

View linked text

This point is less strong. Typically, a writer should only include reference to sources of possible error that would have affected one condition of the experiment differently to other conditions. Noise is likely to have had a similar effect on both Sections 3 and 5 (particularly as the study used counterbalancing).

View linked text

This bring the point to a conclusion which reflects your argument. However, once again the use of literature here would have made the point more convincingly

View linked text

Quality: Authority

This use of literature substantiates your earlier claims and provides a stronger voice for your argument

View linked text

Going on to discuss the literature like this shows that you really do understand it, and are confident in using it

View linked text