a quasi experiment on age and sleep Flashcards
operationalised variables
IV- Age= adolescent (16-24), adult (25+)
DV- Sleep= score on the Pittsburgh Quality of Sleep Index
operationalised experimental hypothesis
there will be an increase in the score on the Pittsburgh sleep quality index in adolescent participants aged 16-24 compared to adult participants aged 25+
directional or non-directional?
directional
explain why you used a directional hypothesis
research on sleep clearly states that adolescents and young adults need more sleep than adults. and due to sleep phase delay in the sleeping patterns of adolescents which is mismatched with school/work, adolescents sleep patterns don’t match their circadian rhythms leading to poor sleep quality. therefore our hypothesis is directional based on this research as we would expect adolescents to have higher scores on the sleep quality index compared to adults.
appropriate null hypothesis
there will be no significant difference in the score on the sleep quality index in adolescent participants aged 16-24 compared to adult participants aged 25+. any difference will be due to chance.
experimental design
independent groups design due to IV being naturally occurring meaning that participants can only belong to one condition
sampling method
opportunity
advantages of sampling technique
more convenient than random sampling as a list of potential participants isn’t needed to select from and instead people who are readily available at the time (bilborough college students and people from the local area) can complete the Pittsburgh sleep quality index which is less time consuming.
As the participant is being approached by the researcher to complete the sleep quality index there are less issues surrounding consent compared to methods such as random sampling where the participant is selected and then could say no when approached by the researcher.
limitations of sampling technique
just because people are available at the time does not mean that they are willing to take part and complete the Pittsburgh sleep quality index.
inevitable bias because the sample is drawn from a small part of the population, will be researcher bias as i am more likely to approach people i know such as friends and family.
description of procedure
firstly, i will approach adolescents and adults in the local area and ask if they are willing to participate in my study. i will make sure to ask that they are over 16 to ensure they are able to take part and provide valid consent. i will then give participants a briefing statement that outlines the study on age and sleep quality, giving details on the sleep quality questionnaire they will need to complete. if the participant is happy with the information in the briefing statement, they will sign the consent form and tick which age group they are in (16-24 or 25+). following this, i will share the link with the participants to the PSQI and ask them to fill this in on their phone. after completion, the participant will share their PSQI score for me to record on my record table. lastly, i will give the participants the debriefing statement outlining the hypotheses and purpose of the study and give them the chance to ask questions or withdraw their data. if they are happy with the debrief they will sign to finish the study. if participants are concerned about the score they received, i will provide them with resources that could help them improve their sleep quality.
measure of central tendency
mean- the mean is appropriate for my data as the DV is measured as score on the PSQI which is a score from 0-21. this is a set scale with a true zero and therefore the data is ratio, making the mean the most appropriate measure of central tendency.
measure of dispersion
standard deviation- as we are able to calculate the mean score on the PSQI for adults and adolescents, the most appropriate measure of dispersion is standard deviation as the mean is required as part of this calculation and gives a more accurate spread of data compared to the range.
appropriate graphical representation
bar chart- displaying the mean scores on the PSQI for adolescents and adults, the data is discrete and therefore a bar chart is most appropriate
inferential statistics
mann whitney u- mann whitney u is the most appropriate inferential stats test for my data as the experiment is looking for a difference in scores on the PSQI between adolescents and adults. the experimental design is independent groups as the participants only complete the PSQI once as age is the IV so participants can only belong to one condition. the level of measurement is at least ordinal as the DV is measured as the score on the PSQI which is a score from 0-21 which is ratio data.
briefly summarise your findings
mean:
16-24= 9.2
25+= 6.6
SD:
16-24= 3.26
25+= 2.88
u value= 28
critical value= 27
conclusions from findings
as the calculated value of 28 is higher than the critical value of 27, the findings were not significant at p=0.05 therefore the null hypothesis was accepted.
issues of reliability and how you dealt with them
as a group in the pilot study, we did not conduct the research together. this means that the way in which we delivered the instructions to each group of participants may not have been consistent between each researcher. this would affect the internal reliability of the study. to deal with this issue, we created a set of standardised instructions on how the PSQI was to be completed, increasing internal reliability.
due to time restraints we were not able to replicate our research on age and sleep quality, meaning we did not assess the external reliability of our study. to deal with this issue, we would do the test retest method. after leaving a time gap of a week we would approach the same participants and ask them to repeat the PSQI. we would then compare the scores on the PSQI from the original data to the new scores a week later. if the scores were consistent we would have good test retest reliability.
how you established your research was reliable
we established that our research was reliable through clear operationalisation of the IV/DV.
another way to establish the reliability of our research is through the split half method. we would split the answers on the PSQI in half (odd questions vs even questions) and compare the scores on each half, if the scores are consistent, then the internal reliability of the questionnaire is high.
issues of validity and how you dealt with them
during our pilot study, we saw that participants were being distracted by other people present as they were discussing their answers on the PSQI. we also noticed that the people we asked to complet the PSQI later in the day were less focused on completing the PSQI accurately. these are extraneous variables that would affect the internal validity of our results. to deal with this we asked each participant to complete the PSQI only in our presence to avoid the distractions of others. we took each person to a quiet area to avoid environmental distractions and we asked them to complete the PSQI around midday to ensure peak concentration, increasing the internal validity of our research.
as the 16-24 and 25+ were answering the PSQI questions, their answers may have been influenced by social desirability bias as they may have felt that their answers about sleep quality may reflect negatively on them. this would affect the internal validity of the study. we dealt with this by only collecting their global PSQI score and did not ask for the scores on specific questions. hopefully this would’ve encouraged them to answer more truthfully as we did not know their actual answers.
how you established your research was valid
content validity- ask a sleep quality expert if the PSQI is an accurate measure of sleep quality
operationalisation- clear operationalisation of age and sleep increases the internal validity of our research
concurrent validity- compare scores on PSQI to another established measure of sleep quality, if both measures give the same results regarding sleep quality, our research could be said to have good concurrent validity
ethical issues and how you dealt with them
as the study is on age and sleep, participants from the local area may feel that it is an invasion of privacy asking their exact age as this can feel very personal and may cause embarrassment. additionally, the nature of the questions may make participants feel uncomfortable due to them being personal. to avoid the invasion of privacy, i only asked which age bracket the participants fell into to avoid asking overly personal questions. i also ensured to tell participants that i will only be using their PSQI score and will not be looking at their individual answers.
as the study consists of participants completing the PSQI to gain a score surrounding their sleep quality it may cause distress if their score is high. to protect participants from harm, i will provide advice and helplines if they are worried about their sleep quality.
2 ways you could improve your research and why they would improve it
more varied/ diverse sample of 16-24 and 25+ as they were all gathered from a small area in the east midlands. i would do this by conducting the study online, as this would allow me to reach a wider and more diverse sample of participants. this would increase the population validity of the study, giving me a more representative result of the effect of age on sleep quality.
i would improve the measure of sleep that we chose to use. the PSQI is a questionaire designed to measure sleep quality but some of the questions were difficult to answer and relied on the memort of participants, such as ‘over the past month how many times have you woken from sleep to use the toilet’. to improve this, i would measure sleep quality through a sleep lab, where a sleep engineer would be able to accurately measure, using EEG scans, the depth and quality of sleep each participant is getting. this would improve the internal validity of my study, as the measure of sleep quality is more accurate.
