Personal investigation two - a correlational study involving a Stroop test Flashcards
Aim
To identify if there is a link between the age of participant and their speed in completing the incongruent Stroop test compared to the congruent Stroop test.
Hypothesis
There will be a relationship between time completing the incongruent Stroop test compared to the congruent Stroop test (in seconds) and age of participant (in years).
Null hypothesis
Age (measured in years) will have no relationship to speed completing incongruent Stroop test compared to congruent Stroop test (time measured in seconds.
Methodology
Correlational study to identify if there is a relationship between age and ability to complete Stroop test.
Methodology - strengths
Allows the research to identify if there is a genuine relationship between the two co-variables which would be impossible to manipulate any other way.
Is more ethical and valid than an experiment as there is no manipulation of the variable.
Methodology - weaknesses
Correlations only identify a relationship between co-variables not a cause and effect. This means that any impact of age on time scored could be due to other factors.
Confounding and extraneous variables cannot be controlled for so any impact on time scored may not always be due to age.
Sample and sampling methods
Opportunity sampling and all participants are in the same group as this is a correlation which simply identifies the relationship between age and speed of completing the incongruent Stroop test.
Sampling - strengths
Opportunity sampling is quick and easy to do and participants were all easily accessible to researchers.
Researchers will be able to access a wider range of ages from the population they have contact with on a day to day basis and they will be a very limited attrition rate due to being asked at the time.
Sampling - weaknesses
Bias within sample as they were all known and were picked purely for their willingness to complete the study.
Limited geographical spread of sample population means there are issues with generalisability to other samples.
Procedure
Each participant was given the consent form to sign.
Researchers had a link to the Stroop test that would also record the time taken to complete the test.
Each participant was instructed to complete the congruent Stroop test – reading the colour words focused on the colour of the word – in this case this matched the actual word. When they were ready the researcher would press to begin and the participant would then read out the words. When all were read out correctly the timer was stopped and the time recorded.
Then the participant went on to the incongruent test reading the colour the word was written in rather than the word itself. The researcher would press the time button to start and the participant would read the list of coloured words out. Once all had been recited correctly the timer was stopped and the time recorded.
All of this information was uploaded onto a Microsoft forms which also acted as the consent form for participants.
Data was then complied taking only the age of participant in years and the difference between scores – taking the congruent time away from the incongruent time.
Data was then analysed using Spearman’s Rank Order Correlation Co-efficient to understand if there was a statistically significant finding from the data.
Issues of validity
Participant variables may have impacted on ability to complete the test.
There may be variability between Stroop tests and methods of recording speed of completion.
Low ecological validity - findings can’t be generalised to the wider population as all participants are from the same area.
Ways we ensured validity
All participants were able to identify if they had any degree of dyslexia or where colour blind.
Ensure that the participants are using the same Stroop test with the same means of timing to ensure that the difference in test score between congruent and incongruent was due to age and not due to errors in timing so had a good level of content validity.
Issues of reliability
Each researcher was tasked with completing the test with at least one participant to ensure that we were able to access a wide range of ages.
Limited age group of participants in the college environment.
Ways we ensured reliability
We ensured that each researcher followed the standardised set of instructions which were a part of the Microsoft form so that any variability of test score was due to the participant and not different instructions or methods of timing from the researchers.
Ensured that researchers picked their participants from a wide range of populations – family, friends, work places etc.
Ethical issues
A slight possibility of suffering from psychological stress if the individual feels they have done significantly worse than they should have.
The participants may be concerned that their name and date of birth was captured in the Microsoft form and that they would be identifiable by this data.
Ways we overcame ethical issues
Participants were able to understand the aims of the study clearly and were also given a contact number if they required any additional support.
The study was done in private locations so any embarrassment was limited to the immediate researcher who was already known to them.
Participants were assured that all their details would remain confidential and only accessible by the lead researcher due to password only access to data and no identifying information would be released. Also made aware of their right to withdraw from the study should they wish to and that all data pertaining to them would be removed.
Statistical analysis
rs = -0.12589, p (2-tailed) = 0.5074.
By normal standards, the association between the two variables would not be considered statistically significant.
Conclusions
This means that for this study we accept the null hypothesis as the observed value of rs = -0.12589, p (2-tailed) = 0.5074. was not equal to or greater than the critical value of 0.306 when N is equal to 30.
This means that there is no significant correlation between age and time difference between the two versions of the Stroop tests. This is also reflected in the Scatter graph shown above which clearly shows little to no evidence of a positive correlation with two very distinct outliers present at two differing age groups.
Overall limitations
Limitations faced in this study are that differing locations, times of day and size of screen may have impacted on the participants ability to complete the Stroop test to the best of their abilities.
An issue of location bias as all participants in the same location.
Overall suggestions for improvement
Researchers could have used a minimum screen size to ensure that all participants had the same ease of access to view and that diminished eyesight was not a confounding variable to include.
The research could take place online with the use of video and camera to ensure that participants are not cheating which would allow access to a wider range of participants and potentially more accurate recording of data as this could be recorded and checked by AI interface.
