Вы находитесь на странице: 1из 5

1) Identify the type of experimental design used in this study [1mark]

Independent measures design


[ the participants in this experiment have all been divided into two groups, where half are in one
condition and other the half in another condition, therefore they are independent]
2) Identify one extraneous variable that the investigator addressed in the procedure for the study
and explain how it was addressed.[4 marks]
One extraneous variable would be individual differences of the participants in the study, where for
example some may be very intelligent and would affect the results. To solve such issue, the
participants were randomly allocated as a way of decreasing the chances of the participants
individual differences having an affect on the results gathered which would consequently affect the
validity of the study. Random allocation could have included the participants being numbered and
then a computer could be used to select which numbers go into which group.
3) Name an appropriate test of statistical significance for analysing this data. Explain why this
would be a suitable test to use. [4 marks]
The Mann-Whitney U test should be used. This is because the experiment involves independent
groups. Additionally the data is ordinal because the scores involved in measuring the number of
ideas can be compared e.g. 5 ideas is more than 4, however it is ordinal as with such comparison
between 4 and 5 ideas, and between 9 and 10 ideas, the measurement of creativity in such way
cant be said to be equal.

[ Any question like, this just remember the table below and work your way backwards to explain the
use of such test]

Nominal
Chi squared test

Differences
Ordinal / Ratio
Independent groups
Mann Whitney
U test

Correlation
Related
Wilcoxon matched
pairs signed ranks
test

Spearmans rank

4) Explain what is meant by p 0.05 [2 marks]


This means that there is less than 5 in 100 [that is with 5 divided by 100 = 0.05] probability that the
results are due to chance. If this occurs then the results are significant with their hypothesis being
supported.
5) Give one reason why the psychologist used a two-tailed test [2 marks]
A two-tailed test is used when the hypothesis is non-directional. As there is no indication that the
research context suggests a direction of difference, a non-directional hypothesis and two-tailed test
would be appropriate.
6) With reference to the data in Table 1 outline and discuss the findings of this investigation. [10
marks]
Using the average number of ideas generated for each it can be seen that those working alone
tended to generate on average more ideas than those working in a group. The average here, most
likely to be the mean of the data, shows that by adding all the data for each group and dividing by
the number of participants in that group, the group where individuals worked alone there were
more individuals that had higher scores than lower scores compared to the other group. The
standard deviation for the condition of working in a group is higher than those working alone. This
means that the number of ideas generated by each participant varied more from the mean for those
working in a group compared to the working alone group where the data for each individual would
have been closer to the mean of 14. In evaluating such findings however, there could have been
possible outliers within the sets of data, where extreme samples would have skewed the data,
where the mean and standard deviation would be particularly vulnerable to the effects of an outlier.
Additionally regarding the task itself there may be an issue with the validity of it because of the
artificiality of testing creativity under such controlled conditions in a laboratory. Alongside this, the
operationalising is difficult where it is hard to define what actually constitutes an idea. Difficulties
arise with reliability where it cant be determined how many people actually participated in
generating ideas in the working a group condition as most of the ideas may have come from the
same individual or some individuals may have not even participated.

7) The psychologist noted that younger participants seemed to generate more ideas than older
participants. Design a study to investigate the relationship between age and ability to generate
ideas. You should include sufficient details to permit replication, for example a hypothesis,
variables, detail of design and procedure, sampling. [12 marks]
A directional hypothesis for this study would be the older the participants are the higher the
number of ideas that will be generated. This hypothesis is one-tailed due to the psychologists
realisation that younger participants generated more ideas than older participants in the previous
study. The null hypothesis would be there will be no differences in the number of ideas generated
between the old and young group and any difference will be due to chance. The independent
variable is the age of the participants, the dependent variable is the number of ideas generated.
Extraneous variables that should be controlled for would be: the environment that the participants
are in, as being in different environments e.g. outside may allow more creativity. To control for this
the participants should be placed in the same environment for each condition, i.e. a room with white
walls, flooring and chairs and no decorations. Another possible extraneous variable would be the
mood of the participants, where if they are particularly emotional they may vary in creativity. To
control for this a closed-question questionnaire could be carried out to determine what mood the
participants are in so this can be factored in to the results.
The sampling for the study would include a volunteer sample. For this advertisements could
be placed in elderly homes and schools to collect the required number of participants, which will be
twenty for each condition. Regarding the experiment itself, a lab experiment will be used as this
offers the benefits of controlling variables in order to determine the effect of age more accurately.
Independent groups design will be used for the experiment with participants being placed according
to age where the Young Group will include participants aged 17-18 and the Elderly Group with
participants aged 70-90.The difference in range between the ages selected is due to the
developmental differences at younger ages being more significant than say the changes that occur
between 70 and 90. Regarding ethics, informed consent should be taken, where for those under 18
parental consent should be given. With this, they should be fully aware of their right to withdraw
and the confidentiality of the information given. Additionally a debriefing should be offered after the
experiment is completed to ensure that participants understand what they have participated in and
do not feel that deception has been used.
[you can determine that the hypothesis is directional as the question states:
The psychologist noted that younger participants seemed to generate more ideas than older
participants.] SEE box below for general outline of what should be included when designing a
study

Designing a study: Questions to be answered when answering these types of question.


Hypothesis should it be one-tailed or two-tailed?
Null hypothesis states that the hypothesis is not true and that the results are due to chance.
Variables: What is the independent and dependent variable? What extraneous variables could arise and
how can these be controlled? (e.g. age of participants, gender, mood, intelligence etc.)
Sampling: What method of sampling should be used? Opportunity, Volunteer or Random?
How large is the sample, will they be allocated randomly to groups?
Method: Lab experiment / Natural experiment / field experiment/ questionnaire / interview /
correlational study / observation
Design: repeated measures / matched participants / independent groups
Ethics: debriefing, informed consent (children parent), confidentiality, right to withdraw, protection
from harm

Some random shit I thought Id throw in:


Peer review independent assessment experts within field they allocate funding
Issues with peer review.
Bias on funding and publication more male scientists than female
Favoured publication of positive results media headlines
Institutional bias e.g. research from higher class universities only published
Encouragement of self-serving groups of reviewers nothing new is permitted
Takes a long time to review
Probability levels:
A probability level p < 0.05 means that there is a less than 5% chance that the results are due to
chance. This level is used because it allows for a reasonable degree of flexibility, but not so much as
to invalidate the results. It is not too lenient as this might lead to a type 1 error, and it is not too
stringent as this might lead to a type 2 error.
A type 1 error is where researchers accept a hypothesis as true when, in fact it is not. It is sometimes
referred to as a false positive. This can occur when the probability level is set too leniently (eg p <
0.1 or a 10% probability that the result is due to chance)

A type 2 error is where researchers reject a hypothesis as being untrue, when in fact it is true. This
can occur when the probability level is set too stringently (eg p < 0.001 or a 0.01% probability that
the result is due to chance)

Вам также может понравиться