Insert your data here – this is the table shown in the instructions document labelled with your student number.
|2107630||Quarter of Year|
Identify the dependent variable
Identify the Independent Variables
|Positions of player
Assess if your data and variables meet the assumptions of Chi-square (Approx. 125words)
|We see the χ2 (sub with symbol) is applicable as the data is nominal, the data is categorised from the two independent variables into groups, no data set can be in more than one group set or lie outside a group set as all participants play football with a certain position and they can only have one birth day so there put into quarters of the years simply. The two variables both have more than the minimum of two needed independent categorial groups. These data are accepted in the contingency table so a fair RAE test can be carried out looking at if the quarter someone is born in effects the position in football.|
Insert the Contingency Table here.
|Quarter of year|
|% within row||31.461 %||25.536 %||24.617 %||18.386 %||100.000 %|
|% within column||38.938 %||38.580 %||40.033 %||38.136 %||38.957 %|
|% of total||12.256 %||9.948 %||9.590 %||7.163 %||38.957 %|
|% within row||28.834 %||25.971 %||24.335 %||20.859 %||100.000 %|
|% within column||17.826 %||19.599 %||19.767 %||21.610 %||19.459 %|
|% of total||5.611 %||5.054 %||4.735 %||4.059 %||19.459 %|
|% within row||31.138 %||27.246 %||21.856 %||19.760 %||100.000 %|
|% within column||13.148 %||14.043 %||12.126 %||13.983 %||13.291 %|
|% of total||4.138 %||3.621 %||2.905 %||2.626 %||13.291 %|
|% within row||33.474 %||25.316 %||23.769 %||17.440 %||100.000 %|
|% within column||30.088 %||27.778 %||28.073 %||26.271 %||28.293 %|
|% of total||9.471 %||7.163 %||6.725 %||4.934 %||28.293 %|
|% within row||31.476 %||25.786 %||23.955 %||18.782 %||100.000 %|
|% within column||100.000 %||100.000 %||100.000 %||100.000 %||100.000 %|
|% of total||31.476 %||25.786 %||23.955 %||18.782 %||100.000 %|
Describe the Pattern of Data from the Percentages (App125 words)
|We see from the percentage of total count most data coming from defenders with all these percentages being greater then 38% in total and compared to goalkeepers with all their quarters it was lower than 14.1% to all other positions. Midfielders there was between 30.09%-26.27% and for forwards 21.61-17.83%, so a downward frequency pattern was highlighted from Defenders to midfielders, forwards and goalkeepers.
The percentages show no significant differences with a person birth quarter and their position in football they showed the same pattern whereas the quarters went from left to right there was a decrease of cells in each position shown as for defenders for Q1 to Q4 the variance (highest-lowest column percentage) was only 1.897% for midfielders 3.817% forwards being 3.784 and goalkeepers 1.917%.
|Quarter of year|
Describe and interpret the pattern of the data from the Observed Values, Expected Values and Standardised Residuals within this table (App125 words)
|From all results there was no significant standardised residuals, as there were no standard residuals having a value greater than 1.96 or lower then -1.96 so little association detected.
The results swing from positive to negative in the expected count with no correlation to certain positions or quarters of the years, the standardised residuals vary from -1.0413 for defenders in Q1 to 1.0595 for forwards in Q4 showing there’s no correlations with the variables, all quarters have positive and negative SR’s and all positions having negative and positive values.
Insert Chi-Square Test table here
Describe and interpret the Chi-Square statistics within the table (App100 words)
|From the degrees of freedom 9 pieces of information can vary without breaking the constraints, we see the probability of belonging to Q1 is statistically significant but that the players position and when they were born didn’t have a significant baring shown in the p value being 0.812 which is higher than 0.05, the chi squared reading (χ2 )is 5.25, so we can reject the null hypothesis if the observed value is equal or larger then 5.25 a sample size of 2513 male professional footballers were collected.
Insert Nominal Measures Here.
Cramer’s V (φc)
|ᵃ Phi coefficient is only available for 2 by 2 contingency Tables|
Describe and interpret the Effect Size statistics within the table (App75 words)
|With the value of φc being 0.026 which is a lot closer to 0 then 1 a non-significant association between the variables is highlighted.|
Report the full result of the Chi-Square test below
|For this data there was a non-significant association between the players positions and there birth quartile χ2 =5.254, φc=0.026, p=0.812, p>0.05 this is shown in figure 1 where the correlation on averages shows little difference, the correlation to quarters and number of players had negative gradient shown in figure 1 from Q1 down to Q4 looking at the study from Delorme (2010) it found the same pattern with a greater number of players born in the first quarter of year (Q1), with these they discussed that the coaches selections were reached attending primarily to the players “anthropometric, physical and physiological variables”, which linked with RAE, as “Q1 players will have these qualities more developed” further evidence in a study from Kearney (2017) on French rugby players showed from 2135 players there tests with the same quarters of the year found here was a over-representation of players born in Q1 (SR=2.66) and an under-representation of players in Q4(-2.88) it also found a link between RAE within forward and back row players, and the RAE is depended on how physically demanding the sport is and this can cause bias in the French rugby system and also for are study on professional football players but that the positions aren’t too depended on RAE as footballers have similar skills in each position. We see the frequencies in are percentages go from the highest for defenders to midfielders then forwards then goalkeepers this was expected with most teams playing in a 4-4-2 or 4-3-3 and some instances 4-5-1 or 5-3-2, there’s always 1 goalkeeper and in most cases more midfielders then forwards and more change of more defenders than midfielders, so the pattern isn’t out of the ordinary.|
Delorme, N., Boiché, J. and Raspaud M. (2010) Relative age effect in elite sports: Methodological bias or real discrimination? European Journal of Sport Science 10(2), 91-96. Page 95
Kearney, P. (2017). Playing position influences the relative age effect in senior rugby union. Science & Sports, 32(2), 114–116. https://doi.org/10.1016/j.scispo.2016.06.009 Page 115
Comments are closed.