Personal Investigations

• Iv - age, DV - sleep, 16-30, 31-60.
• Aim - to investigate how age affects average hours of sleep over 3nights. Directional hypothesis - alternative - older people aged 31-60 will have significantly higher average sleep hour than younger people aged 16-30.
• Null hypothesis - there will be no significant difference in average hours of sleep between younger and older participants 16-30 and 31-60 unless due to chance.
• Independent group design - this reduces the order effects compared to repeated measures where participants could get bored. Location - online - questionnaire to access hard to reach participants. Target population 16-30 years and 31-60 years sampling frame is 16-30 years and 31-60 years in the north east.
• Opportunity sampling, as it is quicker and easier than stratified sampling as can use participants available at the time. Standardised procedures to ensure all participants have the same experience.
• Internal reliability - split half to see if scores are similar, external reliability test re-test to compare scores. May have been researcher bias, demand characteristics, individual differences, double blind/deception, operationalising variables.
• Assessing validity - face validity so whether the test measures what it is supposed to measure. Ecological validity - real setting, real task of sleeping. Population validity - can be generalised to others as a range of participants. Risk of stress, anxiety/humiliation, pain as may feel stressed recording sleep hours right to withdraw given a debrief.
• Results - 16-30 total 107 mean 7.13, range 6.3 31-60 total 104 mean 6.93, range 3.
• Used the mean as all scores in data set. used range as it is quicker and easier to calculate than standard deviation. Used a bar chart as 2 conditions so can easily see the difference between each category.
• Found that 16-30 years on average sleep longer than 31-60 years. Ratio data as a true zero can't sleep less than 0 hours, used Mann Whitney U test as difference, independent group, 2 conditions, ratio, can't sleep less than zero hours so ratio.
• Observed value - 100.5 and critical value 72, probability 0.05, 5%. Observed value > critical value so not significant accept null hypothesis any difference due to chance. Reject alternative hypothesis.
• So younger participants 16-30 years, sleep more than older participants 31-60 years.
• Improvements - 1 - repeat 3 times, test retest. 2 - unrepresentative, sample, can't generalise, should have used stratified not opportunity sampling. 3 - content validity - specialist checks could have asked a sleep specialist about our study.