Personal investigation one - a quasi-experiment on age and sleep

Question 1

Aim

Answer 1

To investigate whether there is a correlation between age and quality of sleep.

Question 2

Hypothesis

Answer 2

The older group ages 40-60 are more likely to score above 7 on the Sleep Quality Assessment which indicates a poorer quality of sleep in comparison to the younger group ages 12-18 who are more likely to score below 7 indicating a better quality of sleep.

Question 3

Null hypothesis

Answer 3

There will be no difference in the results of the Sleep Quality Assessment between the younger age group, ages 12-18, and the older age group, ages 40-60.

Question 4

Methodology

Answer 4

Quasi-experiment – difference study. Independent variable – age.
Dependent variable – sleep quality.
Quantitative data.

Question 5

Evaluation of methodology - strengths

Answer 5

Quantitative data is easier to analyse.
This enables more conclusions to be drawn.
Quantitative data is seen as a scientific and objective way to study variables.
This is a strength as psychologists favour scientific explanations.
A quasi-experiment has less ethical issues than other types of experiments.
This is because the groups already exist.
Quasi-experiments allow researchers to study variables that otherwise may not be ethical or to study, while respecting / adhering to ethical guidelines.
Quasi-experiments have high ecological validity. Quasi-experiments allow researchers to investigate variables in their natural environment.
This means that the findings are more representative of how the two variables would behave with one another outside of a structured laboratory environment.

Question 6

Evaluation of methodology - weaknesses

Answer 6

Since the independent variable is not directly manipulated by the researcher, we can’t be sure that effects on the dependent variable were not due to confounding or extraneous variables.
This makes it more difficult to generalise our findings.
Our participants are all very different from one another, this means that participant variables may affect our findings.
This will also make it harder for us to generalise our findings to the wider population.
Our questionnaire is made up mainly of closed questions.
This may oversimplify reality.
A participant may also feel forced to choose an answer that may not accurately represent how they feel.
This creates the risk that our conclusions may be meaningless if participants were to just select an answer for the sake of not leaving the answer blank, despite not feeling like any of the possible answers relate to them.
Quantitative data doesn’t provide us with detailed information the same way that qualitative data would. To overcome this issue, we made sure to have a variety of questions in our Sleep Quality Assessment and not just limit it to a few questions.
This means that we our questionnaires still provide us with a large amount of information.

Question 7

Sample and sampling method

Answer 7

Opportunity sampling.
20 participants.
The participants were separated into two groups based on age - 10 in the younger age group (age 12-18) and 10 in the older age group (age 40-60).

Question 8

Evaluation of sampling - strengths

Answer 8

Opportunity sampling is the easiest method of sampling since it takes significantly less time to locate your sample.
It therefore allows for a quick collection of data.
We were able to collect large amounts of data within a short period of time.
The size of our sample means that we’re able to gather a large amount of data.
This will help us to establish findings and casual conclusions between our variables - age and sleep quality.

Question 9

Evaluation of sampling - weaknesses

Answer 9

As our participants are self-selected, our sample may be unrepresentative of our target population. Opportunity sampling is also bias as the sample is only drawn from a small part of our wider target population.

Question 10

Procedure

Answer 10

Hand out consent forms to participants.
Once consent forms are completed and signed by the relevant people (a legally responsible adult and the participant for those participants under the age of 16), hand out the Sleep Quality Assessment to all participants.
The participants will then complete the Sleep Quality Assessment, ensuring that all questions are answered to the best of their ability.
The assessment includes questions relating to how long it takes them to fall asleep, how much hours they sleep and any reasons why they woke up during the night.
The questionnaire asks questions about the last month and asks for an estimate for all answers.
The participants then hand the consent form and the Sleep Quality Assessment back to the researcher.
Calculate Sleep Quality score for all participants using the PSQI scoring system.
<7 = good sleep quality.
7-10 – okay sleep quality.
>10 = poor sleep quality.
Add data to frequency table.
Find a mean for both sets of data – younger participants, 12-18, and older participants, 40-60.
Plot two values, the mean, on a bar chart.
Conduct inferential test using Mann-Whitney U to establish whether our results are significant or not.

Question 11

Procedure - findings

Answer 11

The researchers will then come together to calculate the findings of each of the assessments.
We will be using a scoring system almost identical to the one used in the Pittsburgh Sleep Quality Index (PSQI).
The scoring system will leave each participant with a score.
<7 = good sleep quality.
7-10 = ok sleep quality.
>10 = poor sleep quality.
We will place our results into a frequency table and then calculate the mean of each group.
We will plot this mean onto a bar chart.
We will then carry out the Mann-Whitney U test and find out whether or results are significant or not.

Question 12

Issues of validity

Answer 12

Since our participants aren’t randomly selected, our sample may be seen as unrepresentative.
This could affect our results.
Although all our participants are very different from one another, they all live in relatively similar areas.
Our sample is therefore mainly representative of people within the area of Weston.
We don’t know how accurate this sample is of the wider population; therefore, our findings can’t be generalised.
Confounding variables affect the outcome of a study.
Since our study is a quasi-experiment, there’s a significantly higher risk of confounding variables compared to other types of experiments.
Since the variable isn’t manipulated, we can’t prove that it’s age that’s affecting our participants quality of sleep.
This affects the internal validity of our study.

Question 13

Ways we ensured validity

Answer 13

We increased our sample size to get as much of a wide range of participants as possible.
By doubling the sample size, from 5 in each group to 10, it increases the validity of our study.
All of our participants will follow the same set of standardised instructions and procedures.
This ensures that there isn’t any deception or researcher bias.
To limit the impact of confounding variables, we’ve written and included questions in our Sleep Quality Assessment that encompasses the key confounding variables of our study.
For example, medication and stress levels.
By including these questionnaires in our assessment, we’re able to acknowledge any data that may be affected by the variables and apply to this our findings.

Question 14

Issues of reliability

Answer 14

Our questionnaire is self-reported.
Therefore, the data may be biased since people’s memory and perception over the course of the last month may be inaccurate.
As our study was conducted over a short period of time, it’s difficult to guarantee external reliability.
Our research doesn’t prove that the difference between the two variables will be consistent over time.

Question 15

Ways we ensured reliability

Answer 15

We based our Sleep Quality Assessment on the Pittsburgh Sleep Quality Index (PSQI).
This increases the reliability of our study since there’s already a sleep measurement tool similar to ours that works.
The participants will all follow the same set of standardised instructions.
This increases the internal reliability of our study as all participants will be completing identical questionnaires.
To keep the issues with external reliability to a minimum, we conducted our study for as long as possible.
If we were to conduct out study for any longer than a month, people’s perceptions would be even more altered.
This would make our findings less accurate and our conclusions therefore meaningless.

Question 16

Ethical issues - stress

Answer 16

A questionnaire could potentially cause the participant to become stressed.
If you’re asked about something you don’t think about very often, such as your quality of sleep, it may cause you to start over analysing it.
This could have a negative impact on the participant as they may realise their sleep quality isn’t what they wish for it to be.

Question 17

Ethical issues - privacy

Answer 17

There may be an issue with privacy.
Participants may feel that questions about sleep are invasive and something they don’t feel comfortable answering.
The participants are also all people known by the researchers, this may make them feel as if they can’t say no to participating and they have to continue despite feeling uncomfortable.

Question 18

Ethical issues - confidentiality

Answer 18

There may also be an issue with confidentiality.
The participants may feel worried that many people will know about their sleep quality.
If someone feels ashamed or stressed with how poorly they’ve been sleeping lately, they may want to keep this to themselves.
The participants are also all living in relatively similar locations – as all the participants know previously knew at least one of the researchers. Therefore, this may make them feel as if it’s easier for someone to identify who they are.

Question 19

Ethical issues - valid consent

Answer 19

The younger group in a study are ages 12-18. Participants under the age of 16 are unable to give valid consent.
They may not be able to fully understand what the study entails, where the data will be used and any potential risks.

Question 20

Ways we overcame ethical issues - stress

Answer 20

We ensured that provided participants with the relevant contact information for if they did experience any stress.
Our consent form has the contact information for the college welfare team and the Samaritans.
This gives participants multiple ways to reach out for support if they feel they need it.

Question 21

Ways we overcame ethical issues - right to withdraw

Answer 21

Our consent form makes it clear that if a participant is to feel uncomfortable at any point, they have the right to withdraw.
They also have the right to withdraw after the questionnaire is completed and their results will no longer be used. T
his ensures that the participants don’t feel as if there privacy is being invaded, and if they do, they don’t have to continue.

Question 22

Ways we overcame ethical issues - confidentiality

Answer 22

We ensured that the consent form laid out, in detail, exactly which groups of people would be told about and discussing the participants quality of sleep.
Our consent form states that, “Weston College A-Level Psychology (group A) would be discussing the data collected within the college setting”.
We will also not be sharing any personal information about the participants, (other than their age since this is a key part of the study), as we want to ensure that none of the participants identities are revealed.

Question 23

Ways we overcame ethical issues - valid consent

Answer 23

To overcome any potential issues with valid consent, for any of our participants under the age of 16 we’ve ensured that we’ve got valid consent from not just them but also a legally responsible adult.
We made a second consent form, following the same formatting and standardised content as our first consent form, for adults to sign for their child / relative.
Getting consent from the adult and the child is crucial in making sure that there are no issues regarding consent in our study.

Question 24

Ways we overcame ethical issues - deception

Answer 24

There was no deception in our study.
All the relevant information is set out on the consent form and on the top of the Sleep Quality Assessment in simple sentences.
This decreases the likelihood of anything being misinterpreted.
We also didn’t need to debrief our participants after the study and reveal the true aims as our aims were made clear from the start.

Question 25

Findings - younger group

Answer 25

4
11
12
10
9
5
3
7
4
9

Question 26

Findings - younger group - mean

Answer 26

7.4

Question 27

Findings - older group

Answer 27

9
9
6
7
10
3
7
12
16
13

Question 28

Findings - older group - mean

Answer 28

9.2

Question 29

Findings - summary

Answer 29

The younger age group scored much higher than we originally anticipated. We used the same scoring system as the PSQI, and we found that our scores varied drastically from one another.
We expected the scores of our younger group, ages 12-18, would be significantly lower than those of the older group, ages 40-60.
However, we found that everyone’s scores were very different, regardless of what age group they were apart of.
For example, both groups had a participant who scored 3 on our Sleep Quality Assessment.

Question 30

Mann-Whitney U Test

Answer 30

U value – 37.5.
The critical value of U at p <0.5 is 23.
Therefore, the result is not significant at p <0.5.
Z value - -0.90711.
The p value is .36282.

Question 31

Mann-Whitney U Test - conclusions

Answer 31

Therefore, we reject the experimental hypothesis as the observed value of 37.5 doesn’t meet the criteria of equal to or less than the critical value of 23.

Question 32

Conclusions

Answer 32

Our findings suggest that our participants all had varied sleep quality.
It doesn’t solely suggest that the older the participant, the poorer their Sleep Quality.
Some of our findings indicate this, however some of our results from the younger group also indicate poor sleep quality. Our standard deviation score of 13.2288 supports this, with our results showing worse sleep quality among all of our participants.
We therefore have concluded that it is other factors that impact sleep quality more than age.
Age has an effect on an individual’s quality of sleep; however, this is in combination with confounding and participant variables such as stress levels, living situations etc.

Question 33

Overall limitations - size of the two groups

Answer 33

The two groups in our study don’t have an equal age gap between them. The range of the younger group is 6, whereas the range of the older group is 20.
This means that the data collected from the younger group is much more limited.
It also means that our scores are much more varied than we hoped. Both factors combined reduces the accuracy of our findings means that we have reduced strong evidence to support our hypothesis.

Question 34

Overall limitations - scoring system

Answer 34

We calculated our scores using the Pittsburgh Sleep Quality Index (PSQI) scoring system.
Although this was successful and it made our results much more in-depth than they otherwise would have been, it was complicated.
It’s much easier to get confused with the PSQI scoring system compared to other, simpler, methods of scoring.
This reduces the overall reliability of our findings since a mistake in calculation could easily have been missed.
Making any mistakes in our calculations would make our conclusions meaningless.

Question 35

Overall suggestions for improvement

Answer 35

If we were to repeat this study, it would be advantageous for us to collect our sample from outside of the Weston area.
If we were to gather participants from across the country, it would reduce location bias and increase the ability for us to generalise our findings.
If we were to randomly select our participants, using sampling methods such as volunteer sampling or random sampling, it would reduce any potential selection bias.
It would also mean that we’d be able to have a larger sample size, increasing the validity of our study.