Influence of Gender stereotypes - are males more accurate at giving directions than females?
Title
Influence of Gender stereotypes - are males are more accurate at giving directions than females
Experimental Hypothesis
In the condition where the direction-giver was male, participants in this study would judge that the driver would follow the directions for longer than in the condition where the direction-giver was female. This is a one-tailed hypothesis as a direction is predicted.
Abstract
Based on the theoretical work of Darley and Gross (1983) and Tajfel (1981) the current study sought to investigate how participants judge the accuracy of directions based on whether the direction-giver is either male or female. Additionally, this study predicts that if the direction-giver is male the driver would deem these directions to be more accurate than if the direction-giver is female. This study employs a "between participants design" consisting of 20 participants, 10 males and 10 females. The participants were instructed to listen to a short story and to answer a question. The results indicate that there is no difference between groups whether the direction giver is male or female. Therefore one conclusion drawn is that future studies may want to have selection criteria for participants so as to control for confounding variables.
Method
Design
A between participants design was used in this experiment. The Independent Variable was the gender of the person giving directions, either male or female. This produced two conditions. Condition 1 (experimental condition) consisted of 10 participants, 5 males and 5 females and the gender of the direction-giver was female. A short story (vignette) describing a driver asking for directions was read out loud to the participants and they were then asked to state the number of miles they would continue to drive before taking an alternative action. Condition 2 (control condition) was exactly the same as condition 1 with an equal number of men and women, the only difference was that the gender of the direction-giver in this condition was male. The dependent variable was the number miles the participant stated they would continue to drive before taking another course of action. The participants were asked to provide a single number of miles rather than a range (see full instructions, Appendix 1). The data type was continuous and was recorded on a response sheet.
To control for any potential confounding variables, such as individual differences, participants were randomly assigned to either one condition or the other. However only six out of the twenty participants were recruited by the researcher. While their age, gender and response was provided, their background was completely unknown. There could be a number of unknown variables that could confound the results for example all the participants in one condition could be Spanish speaking or all non-drivers could be in one condition. Demand characteristics where participants respond as they think you want them to and not spontaneously are limited ...
This is a preview of the whole essay
To control for any potential confounding variables, such as individual differences, participants were randomly assigned to either one condition or the other. However only six out of the twenty participants were recruited by the researcher. While their age, gender and response was provided, their background was completely unknown. There could be a number of unknown variables that could confound the results for example all the participants in one condition could be Spanish speaking or all non-drivers could be in one condition. Demand characteristics where participants respond as they think you want them to and not spontaneously are limited in between participants design. The six participants recruited by the researcher were given the same instructions objectively and were tested individually in a quiet environment during the hours of 10am - 5pm.
Participants
The majority of participants (14 out of 20) were not recruited by the researcher and their background was unknown. The remaining participants were colleagues and friends who agreed to take part in this study. There was no reward offered and they were not aware of the hypothesis or knowledgeable in the area of research. These participants had completed a third level education with English as their first language. The age range for all 20 participants was from 18 to 58, with the ratio of men and women being equal across conditions.
Materials
There were two sets of stimuli (vignettes), one labelled condition 1 and the other labelled condition 2. Each participant was allocated to one condition only. The "vignette technique" was used which is a short description of a fictitious event used in experiments to which research participants are asked to respond. In this study the vignette (approx 200 words) was read aloud to each participant. The general gist of the vignette was about a driver having trouble finding a hotel. Having driven around for more than half an hour the driver stopped and asked for directions. In condition 1 the gender of the direction giver was female and in condition 2 the direction-giver was male. The direction giver knew the hotel although couldn't be specific about the distance. (See appendix 1 for vignettes used). A response sheet was used to record the age, sex and the response of each participant (see response sheet, Appendix 3)
Procedure
The participants recruited by the researcher were approached and asked if they would be interested in taking part in an experimental study which required them to listen to a short story that would take approximately 2 to 3 minutes (this was timed in advance) and then to answer a question. The age and sex of interested participants was recorded on a response sheet. The instructions given to each participant were exactly the same (full copy of instructions in Appendix 1). Participants were seen individually in a quiet environment between the hours of 10am and 5pm. Having ascertained that they understood the instructions, a short story (vignette) was read aloud. Each participant was read only one condition. At the end of the story each participant was asked how many more miles would they would continue to drive before taking an alternative route and to give a single number of miles rather than a range. The experimenter recorded the number of miles on a response sheet. The participant was then debriefed and given an opportunity to ask any questions regarding the experiment.
Results
The experimental hypothesis for this experiment was that in the condition where the source of person giving directions was male, participants would judge that the driver would follow the directions for longer than in the condition where the direction-giver was female. This is a one tailed hypothesis as it predicts a direction. The data collected was a single number of miles and their response was recorded on a response sheet. Descriptive statistics for this data and are shown below.
Mean number of miles
Condition
Mean
Standard Deviation
Condition 1
4.80
4.41
Condition 2
5.00
4.19
The mean number of miles for condition 1 where the direction giver was female is slightly higher than that for condition 2 where the direction giver was male. These results indicated that there was only a small difference between these conditions which suggest that the gender of the direction giver did not influence the number of miles the participants claimed they would continue to drive.
Additionally, an error bar chart (see below) showing a considerable overlap between the confident interval levels of both conditions.
Graph
Error bar chart showing confidence interval levels for TMA03. Exp. 2
Condition one - Upper bound confidence interval level - 7.96
(Experimental group) Lower bound confidence interval level - 1.64.
Condition two Upper bound confidence Interval level - 8.00
(Control group) Lower bound confidence Interval level - 2.00.
This suggests that both samples are from the same population. These results did not support our hypothesis, however, more analysis were carried out to determine if this result was statistically significant. For this purpose an Independent-Samples t-test was conducted (See Appendix 2 for SPSS output). The results showed that there was not a statistically significant difference between the two conditions whether the gender of the person giving directions was male or female (t=-.104, df=18, p>.05). Based on this result we have to accept the null hypothesis.
Discussion
As can be seen from the results the difference between the mean scores is not statistically significant. The number of miles that the participants said they would continue to drive was no different whether the gender of the direction giver was male or female. Therefore we have to reject the experimental hypothesis. This non-significant result could be due to the fact that the background of the majority of participants (14 out of 20) in this study was unknown.
It is important to note that when the experiment was completed the participants all said that they were not influenced by the gender of the direction-giver when deciding on the number of miles they would continue to drive. It could be that peoples gender stereotypes do not include stereotypes for giving directions. In this particular study it is important to emphasize that participants received directions from only one male or one female therefore, this study does not attempt to make generalisations.
Findings from previous empirical research have shown evidence of biases effecting participants judgement for example in a study carried out by Darley and Gross (1983, cited by Phoenix, A. 2002). In this study, information about a woman's socio-economic background influenced participant's judgements about her academic ability. One explanation for this difference could be that direction giving is not a trigger for schemas. Schemas are shared knowledge and contain general information which get activated when certain information is presented (Phoenix, A 2002). Direction-giving may not be as salient a topic as the influence of ones background on academic ability.
Based of the body of empirical research carried out previously, such as Henri Tajfel (1981, cited by Phoenix, A 2002) who developed the social identity theory) and postulates that people are judged by their group membership and the Darley & Gross study (1983, cited by Phoenix, A 2002) regarding stereotypic beliefs and how these effect peoples judgement, it is quite likely that there were some methodological flaws in the current study such as confounding variables due to the lack of background information for the majority of the participants.
Future replications of this experiment would need to have basic selection criteria for example all participants need to speak English, receive the same instructions and be similar to each other so as to limit the effect of individual differences.
It could also be argued that experimental social studies do not accurately reflect real-life situations. The general issue of ecological validity is important to reflect on as reading a vignette is hardly comparable to experiencing a situation in everyday life.
Another point to note for future studies would be to increase the sample size as this could help to overcome individual differences.
References
Phoenix, A (2002) Identities and Diversities, in Miell, D., Phoenix, A. and Thomas, K. (eds) Mapping Psychology, Milton Keynes, The Open University.
Phoenix, A. (2002) Understanding the Social World, in Miell, D., Phoenix, A. and Thomas, K. (eds) Mapping Psychology, Milton Keynes, The Open University.
- 1 -
