Both these experiments by Sherif and Asch show that people will go with a group norm and conform to other people who they might necessarily not even know.
Rationale
The experimenter is interested in conformity in contemporary Britain today. The reason for this is because conformity occurs in everyday situations and can have a negative and positive affect on a person’s behaviour. The affects can be great like going along with robbing someone even when you know its wrong, or small like going to the cinema even though you don’t want to but the rest of your friends do.
For background knowledge the experimenter will focus on the Asch experiment carried out in 1952 for help in explaining conformity. Even though this experiment was carried out nearly half a century ago it explains conformity well and also it will be interesting to focus on any difference in conformity levels nowadays. Studies like the Asch experiment (as already referred to in the background knowledge) which was carried out half a century ago, shows a high number of people willing to conform with others, it will be interesting to see if there is a large amount of people who conform with others nowadays
Aims
To investigate if there is higher tendency for males to conform more than females
Hypothesis
Null:
Any difference in conformity levels between boys and girls are due to chance factors
Alternative:
Males will conform more when answering the questionnaires more than what females will
Method
Design
I have decided to use the research method of questionnaires in this study of conformity because they are fairly easy to set up, and most people are willing to participate in filling in a questionnaire as they do not take up much time and are easy to complete. Unlike a laboratory experiment where you have to get your participants into a laboratory where it will take more time to complete.
Variables
The only thing that will change throughout this investigation is the gender of the participants, with 10 males and 10 females being asked. When we’ve collected all 20 questionnaires in I will work out how many questions were conformed in and therefore that’s what will be measured.
Participants
I will use random sampling, as I haven’t got the time or resources to ask participants that are readily available. The participants that I will use will include people from school, family and friends. I will out of these people obtain 10 males and 10 females.
Apparatus
The experimenter will use fifteen ‘Who wants to be a millionaire’ questions to be put into the questionnaire. The questions will be on a variety of subjects so the general knowledge of the participant is tested on a variety of subjects. A computer will be needed to type up and print out the questionnaires, making the questionnaire clear and readable for the participants. (Refer to questionnaire in appendix) Obviously for the study to exist 20 participants, 10 male and 10 female will have to be used. The way in which we will measure the conformity levels of the questionnaires is by seeing out of those questions that were wrong on purpose in the false questionnaires how many people conformed to those wrong answers. Therefore the levels of conformity will be out of ten as there are 5 answers in the false questionnaires that are right answers while the other ten questions are the wrong answers in the false questionnaires and therefore we will see how many of those questions are conformed in. We will give the participant a blank questionnaire as well as three ‘false’ questionnaires.
Procedure
Here is a step by step guide to carrying out this investigation into conformity:
- Get together 15 general knowledge questions. Maybe from a board game that already contains general knowledge questions, or use your own knowledge for the questions. However, the questions cannot be too hard or too easy.
- Type up questionnaire so that it is presentable and readable for the participant’s
- Print off 23 questionnaires
- ON three of the questionnaires circle 10 wrong answers to 10 questions and on the other 5 questions circle the right answers. Make sure you circle the same answers on each of the 3 ‘false questionnaires’.
- For each participant give them a blank questionnaire and 3 false questionnaires, telling them that the other 3 questionnaires are there for them to conform or discard any answer that they think is right.
- Tell the participant that they will have a maximum of five minutes to fill the blank questionnaire in
- All together 20 participants should be asked, 10 males and 10 females.
- After all participants have filled ion the questionnaires analyse them to see how many of those 10 wrong answers in the false questionnaires were conformed in, seeing if there are any differences in the conformity levels between males and females.
Controls
Maybe it will be just a coincidence that the answers that the participants put down correspond to those answers that are wrong in the false questionnaires; therefore no concrete conclusion of the conformity levels can be made. Also the various ages of the participants may affect the results as older people have a wider general knowledge than younger people and will have more of an idea of what the correct answer is and as a result not conform with those wrong answers in the false questionnaires.
Before each participant fills out the questionnaire we will make sure that they consent in doing a psychology study, tell them that they can withdraw at any time (protecting them from stress) while doing the questionnaire and also notify the participants that there answers they put will stay confidential. Also after the participant has filled in their questionnaire I will debrief the participant and explain what the study will show. Doing these things will make sure that the study complies with ethical guidelines.
Making sure that the amount of males and females are kept constant is the only control that we can really focus on. Therefore other factors may affect the levels of conformity e.g. the environment in which the questionnaire is filled in, which I will discuss when explaining the validity and reliability of this study.
Results
Summary table
As you can see by looking at this summary table most of the averages worked out show that overall males have the tendency to conform more in the questions than females. However this is only a slight difference (2%).
Conformed in
Not conformed in
Descriptive statistics
By referring to the summary table you can see that by looking at the various averages shown i.e. mean, median and range males tend to conform slightly more than females. By looking at the result for the mode it seems that question 6 (refer to questionnaire in appendix) was the question that both males and females conformed with. Also by referring to the graph you can see that males tend to conform more than females as the ‘conformed with’ bar is bigger slightly than the females bar. To be precise males conformed 2% more with the false questionnaires than what the females did.
Relationship to hypothesis
By referring to the graph you can see that males conform slightly more than what females do. That would relate to the alternative hypothesis. However, the levels in conformity are very small, which could obviously be to do with chance factors. So it seems that the best hypothesis to use for this set of data is the null hypothesis, ‘Any difference in conformity levels between males and females is down to chance’.
Discussion
Validity of results
Seeing if the results actually measure conformity in an efficient way can conclude the validity of the questionnaires. There fore referring to the method and results it shows that this research method of questionnaires lacks validity. There could be various reasons for this. One of these reasons is that maybe the participants did not actually conform by looking at the ‘fake’ questionnaires. Instead just out of coincidence with the study circled the answer, which they thought, was genuinely correct, and it was just out of chance the same answer had been circled in the false questionnaires. As a result participants who may have put six wrong answers down in their questionnaire may not have necessarily conformed but just got the wrong answer, therefore the results would not be valid.
Even though conformity is an everyday occurrence and that the results show that people do conform the research method of questionnaires does lack mundane realism. The reason for this is because receiving a questionnaire with three other questionnaires, which you can look at, does not really happen when you fill in a questionnaire. As a result participants may guess what the questionnaires were setting out to find out, and therefore not take any notice of the false questionnaires. Obviously this is another reason why this research method lacks validity.
How to improve validity
The only real way of improving validity in this study is to change the research method totally. The reason for this is because questionnaires only really test informational conformity not behavioural conformity. Therefore having a research method like Asch’s study would improve the validity of this study a lot. This would also comply more with real life and then the study would be a lot more valid and respectable than just testing informational conformity.
Reliability
This research method of questionnaire lacks reliability for various reasons, as the extent of what the questionnaires measure does not really refer to measuring conformity consistently.
Reasons for this lack of reliability include the lack of a standardised procedure. None of the participants were given certain set instructions; therefore each participant was treated in a different way. The environment varied a great deal between participants again influencing the reliability of the results. Some participants were asked to fill in the questionnaire in a busy playground with lots of children running around, which could obviously effect concentration levels when answering the questionnaire. On the other hand some participants were asked to fill in the questionnaire when they had plenty of time on their hands e.g. when watching T.V. These participants would no doubt be able to concentrate more when filling in the questionnaire. Also a lack of reliability may be caused by there being a lot of differences between each participant. The only thing changed between the participants were the gender, no account of origin, ethics or ages were involved in the study. Differences in age could mean that because the older people have lived longer they will have a wider range of knowledge compared to younger people, therefore older people may conform less because they may know the answers better. Because of this if the male participants were of a younger age than the females it wouldn’t be a surprise if the females conformed less than the males because they have a wider range of general knowledge, which could cause the females to conform less.
How to improve reliability
By having a standardised procedure i.e. giving each participant the same instructions, same time to fill in the questionnaires would improve reliability because each participant would be under similar controls to the next participant. If the environment were the same for each participant e.g. sitting each participant in the same room at similar times of day would also increase the reliability of the study, as each participant will be able to concentrate at the same level as another participant.
By controlling the age of the participants, for example only asking participants of a certain age group e.g. 30 to 40, 10 males and 10 females will cause there to be more reliability as you will be able to control the amount of general knowledge between participants better. As a result compare the difference in conformity levels between males and females better.
Implications of the study
The results of the experimenters’ questionnaires show that people are willing to conform on something they don’t know about rather than just use an intelligent guess. Between males and females there isn’t much difference in conformity levels, even though the results show that males do conform slightly more than what females do, but the result is too close that you cannot make a concrete conclusion about any difference in conformity levels between males and females.
Overall you can say from this study in contemporary Britain people of both genders are willing to conform more than what they would by just going with their own opinions.
You can’t really say from these results that they correspond to the same conclusions that Asch and Sheriff gained form their studies, because their studies showed a lot more significance about conformity than this one. Also different types of conformity were tested in the Sheriff and Asch studies so no comparison can really be made.
Generalisability
To a certain extent it can be said that the results form the study show that people these days will conform more than just giving their own opinions. However, the experimenter only used 20 participants, which is a small amount amount of people to make a generalised conclusion about, as individual differences could play a big part in the outcome of the results. Also the results saying if there is any difference in conformity levels between males and females is too close to make a generalised conclusion, therefore any difference in conformity levels between males and females if due to chance.
Real life application
Even though the research method of questionnaires lacks mundane realism the actual concept of conformity is an everyday occurrence. Conformity can have little effect e.g. going along with your friend to the cinema even though you don’t really want to, or a big effect, for example during World War II German when complied with Hitler to kill many Jews even though they knew it was wrong. Both effects, large or small, can be to some extent linked to the results, as it shows that people do conform to one another. Therefore no matter how big the effect of conformity people life’s can be affected by the fact that they are willing to conform more than going along to what they feel themselves.
References
Asch – Mastering Psychology by Roger Davies and Peter Houghton
Sheriff – Theoretical Approaches in Psychology - Matt Jarvis