Personal investigation 1

title

A content analysis of post match interviews

What is content analysis

a research method where you study artefacts other people have produced and look for patterns in them

• type of indirect observation (not observing behaviour as it happens, looking at evidence after it has been recorded)

how is content analysis carried out

1 choose what to study e.g YouTube videos

2 select sample e.g how many interviews, from who and when

3 create a coding system - categories to sort the data into e.g internal and external attributions

4 analyse the content - record when a category appears

• →if you count how often its quantitative data

• → if you describe its qualitative

5 compare results - look for patterns or differences

what has previous research on internal/ external attributions shown

• athletes often show a self-serving bias - they tend to explain success using internal factors, explain failure using external factors

• Biddle (1988) - race winners were more likely to explain success using internal factors like effort, losers were more likely to use external factors

aim of research

to investigate whether players use more internal attributions after wins and more external attributions after losses in post-match interviews

what hypothesis have we used and why

• directional hypothesis

• as there is previous research e.g by Biddle that suggests a clear expected pattern that athletes show a self-serving bias where they explain their success using internal attribution and failure using eternal attributions

what does operationalised variables mean

• the variables are clearly defined and made measurable in the study

what are the 2 variables that are operationalised

match outcome - whether the player won or lost

• operationalised using the final score - clear and measurable

type and frequency of attribution - how many internal or external attributions are made during each interview

• operationalised through the use of a coding system which records and counts each attribution

your operationalised directional hypothesis

• Payers in post match interviews following a win will make a higher frequency of internal attributions (e.g effort or skill)

• while players in post match interviews following a loss with make a higher frequency of external attributions (e.g luck or conditions)

• as measured using a coding system that tallies each attribution type

your operationalised null hypothesis

• there will be no significant difference in the frequency of internal and external attributions made in post-match interviews following wins or loses

• as measured using a coding system that tallies each attribution type

• any differences will be due to chance

strengths of this content analysis

• good mundane realism as real interviews are used

• study can be replicated through same interviews and coding system

weakness of content analysis

• researchers may interpret statements differently reducing reliability

• researcher cannot ask follow up questions

what was the sampling method used any why

opportunity sampling

• selected interviews that were easily available on YouTube and met the criteria for the study

• practical, quick and convenient

limitation to the oppurtunity sampling

• may not represent all athletes or sports

• reduces generalisability

what descriptive statistics (measures of central tendency and dispersion) were used and why

• mean - allows the researcher to calculate the average number of internal and external attributions in each condition

• range - shows the spread of scores and whether some interviews contain more attribution that others e.g interviews range from 2-9 attributions

what graphical representation was used and why

• bar chart - the data is categorical and based on frequencies

• allows easy comparison between internal and external attribution in wins and loses

what inferential statistic test was used and why

• Chi- squared:

• the data is nominal (named categories e.g win, loss and external/internal)

• testing for a difference

• data is unrelated as there is only one condition

the method used to assess reliability and how

• inter-rater reliability - consistency between two researchers

• all students using the same standardised coding system to code the same first 5 clips and then comparing their tallies

• one teacher provided a loom video to demonstrate how to carry out content analysis so all students follow the same procedure and apply the coding system in the same way

issues with reliability an how we would deal with these

• different researchers may interpret the same statement differently

→ can be reduced by using clear, operationalised definitions and examples of internal and external attributions

• interviews vary in length - longer interviews have more opportunities for attributions

→ only watching a fixed time segment (2 minutes) of each interview

Methods chosen to assess validity

• Face validity - looks like it measures what its spot to

→ getting teachers or students to check whether the categories represent internal and external attributions

• construct validity - how well its measuring the concept its spot to measure

→ ensuring that the coding system is based on attribution theory and previous research into self-serving bias - based it on the coding systems from previous research

issues with validity and how we dealt with them

• coding categories may not accurately represent internal and external attributions

→ coding system is based on attribution theory and previous research

• coding system may appear unclear or inappropriate to other people

→ students and teachers to review it to see if they clearly represent int and ext att or carry out pilot study to identify unclear categories and improve them

how did we deal with ethical issues

• valid consent - YouTube videos don’t require consent as they are publicly available, doesn’t involve direct participants

• risk to participants values, beliefs, relationships - players treated with respect, not mocked

procedure

• teacher will provide 5 pre-selected post match interviews from a range of sports including both wins and loses to assess inter-rater reliability

• additional interviews selected by the student to reach a total of 10: 5 wins and 5 loses

• record the interviews links, player names, player outcomes and dates

• use the pre-prepared coding sheet and behavioural categories for internal and external attributions

• watch each interview once to gain an understanding

• watch the interview again and tally each internal and external attribution

• do not code unclear or ambiguous statements

• to ensure consistency watch the Loom video demonstration which models how to carry out content analysis

• add up the totals for each category and calculate the mean and range

• present the results using a bar chart

• use Chi-squared to analyse whether attribution the differs based on match outcome