After careful consideration over the possible investigations that can be carried out with the data at hand, I have made a decision on the investigation that I am going to carry out. I want to investigate the times in ten minute periods at which goals are scored; this will be done for all four of the league results that I have. I want to do this investigation so that I can find out in what ten minutes of the game, the most goals are scored.
The way in which I can obtain the data that I want to analyse, has to be unbiased. The reason is so that the investigation becomes fair and as correct as it can possibly be. I already have me resource for data, this resource contains the football results from four of England’s largest football leagues and contains the results of them for six consecutive weeks. I am going to use this resource to carry out my investigation; these results are totally factual and have no biased approach to my investigation whatsoever.
From observation of the resource that I have, I see that the Premiership league has less games played on a weekly basis than the other leagues do, this is due to less teams being in the Premiership than in the other leagues and divisions. Due to this, I will have to pick as many teams as possible per week from the Premiership for analysis, and then use that same number to pick from the other leagues. For example, if the Premiership only has nine results and the other leagues have eleven, then to make the investigation fair, I will pick the maximum number of results from the Premiership, that would be nine, and then I would pick nine from the other leagues and divisions.
The main thing to do at this stage is to choose the nine results from the other leagues totally at random so that the selection is not biased in any way. To make things as simple as possible, I will use a method of sampling called clustering. I will pick a random starting point on the other leagues and then pick the nine results that come after that starting point. Of course this is just an example and therefore with the real investigation, the number may be different, so until then this is the method that I will use for the investigation. I will carry out this method for all of the leagues and divisions then take the goals that were scored and at the times that they were scored, then put them in a table which will be split up into ten minute periods. I will then analyse the results that the table shows and hopefully draw up a conclusion on the time at which the most goals are scored.
During the investigation I will probably make new discoveries and findings. After this occasional new finding, I may add a not describing how I am going to add a new sub investigation to the main project. The sub investigation(s) will be added so that the answer to it or the answers to them can be used to help me find the answer to the main investigation; it is a support method to help with the investigation.
There are many possible sub investigations that can be pointed out already at this stage. The reason why I have not stated them in simple. The reason is because I do not know whether the sub investigation would be relevant in aiding the correct conclusion to my main investigation. Due to this, I cannot state the sub investigation now, however, later on during the investigation; I will be able to get a clearer image of the sub investigations in mind and whether or not they will actually be relevant for answering the main investigation clearly.
The Pilot Study is just like a preliminary experiment, this will be in aid for the collection of data that will take place during the main investigation. What I mean by this is, the pilot study will help me to understand how to efficiently collect the correct data and how to do it. The study will also include analysis of the data which helps me to prepare for the real data collection and analysis. Through the pilot study, I can get a good insight on the actual investigation. The results from the pilot study, after careful analysis, will be able to give me a basis for my prediction on the final conclusion to the main investigation.
The pilot study that I am going to carry out will be of the same sort that has been planned for the main investigation. I will take results from the Premiership League and the Nationwide Division One from the fifth week of fixtures and results. I will take four completely random results from each of the selected league tables and then analyse them on the same basis/question as the main investigation.
I picked the two leagues in a sort of clustered sample, however you could just say I picked them for quick data collection rather than having to go from one league to another, these leagues are next to each other in the resource book making the data collection that bit easier for me. Since this is just a pilot study, I only wish to get the gist of the main investigation, and that is why I have not spent so long on the selection.
Otherwise, I would have probably taken the data at hand into careful consideration and then picked the right data for the job. I would have also picked more data, as I can predict that four from each of the two leagues, would not have given me any sufficient foundation to have based a conclusion on, and therefore this pilot study will only help to make a prediction for the main investigation. When the time comes for me to carry out the real investigation, then I will surely use a wider variety of results, and would include a larger number of results in my investigation. This will be an advantage since I will be able to draw up a correct conclusion.
PILOT STUDY RESULTS
The previous table and its results show the goals scored at the ten minute periods within the ninety minutes of the football match. Each period of ten minutes has a number of goals that were scored in that period, the number of goals is shown for the Premiership, and the Nationwide Division One, and it also shows the overall number of goals that were scored in the match at the certain periods of time. The overall number comes from the combination of the Premiership goals and the Nationwide Division One goals put together to give an overall total for the number of goals scored in that period of time.
The reason for creating the table in the manner that I did was to see if there is any difference between the two leagues and the times that the goals were scored. The times at which the goals were scored could show a difference or a relationship, and this is what I want to find out to help me with the main investigation. I would also want to check how many goals overall were scored and how many goals were the difference at each ten minute period. For example, I would want to see if at any one ten minute period, the number of goals scored in each league is different. This would show me what league provides more goals in its games. I will be looking for any relationships between the two leagues and for any major differences between the two leagues and then make any sub investigations for them.
However, looking back at the table of results, I can already see that there are not enough results to see a certain conclusion which could come from the analysis of the results. Due to this reason, I will be thinking of creating another table exactly the same as this one, however, I will take as many results as I can from the two leagues, but most probably from two weeks rather than one. I will select two weeks at random, and the reason for having two weeks to take results from would be simply because I know that more results are needed for me to draw up a satisfactory conclusion on in my pilot study. From this conclusion, I will be able to make a solid prediction on the main investigation. I think that this new table of results and new number of results will help me to see any relationships of trends etc. This will help me draw up a suitable prediction for the main investigation.
PILOT STUDY ANALYSIS
From the table of results from my pilot study, I have stated previously, I can tell that there is not a sufficient number of results to decipher any solid trends or relationships. However, from the table I can see one ten minute period that brings the most number of goals in the match. Within the last ten minute period of the game, I can see from the overall results that the most number of goals are scored then. This I have taken from the table of results that I drew up for my pilot study. Since this is the overall number of goals which are scored, it means that the goals come from both of the leagues that I have taken results from, which are of course the Premiership and Division One.
The goal results put together for week Five out of the eight in total fixture results that I have analysed, together combined, I have found that the number of goals that come in the last ten minute period overall is the most in whole pilot analysis. Since the most number of goals are scored in the last ten minutes, I would now have to analyse what league brought the most goals out of the two leagues in the study and whether the rule of goals coming the most in the last ten minutes applies for the Premiership only or Division One, or even for both.
On analysis via the table of results that I drawn up, the results show that the Nationwide Division One only follows the rule that was set which made out that the most number of goals coming in a match would happen in the last ten minutes of the game. This can be seen because three goals are scored in the last ten minute period overall out of the matches analysed in Division One. From this, I can see that the largest amount is scored then, the reason being that the largest number of goals scored on the tally chart shows that three goals is the largest number whereas the other goals scored number is either no goals scored, one goal scored, or two goals scored, meaning that the three goals scored is the highest number of goals scored overall, this shows that the rule applies for this league. This shows clearly that since there is only one ten minute period which shows three goals being scored overall, that that particular ten minute period was the period that has the most goals scored. This rule that has been set, only accounts for the Division One. I have seen this from analysing the results which show that three goals come altogether in that last ten minute period.
However, in the Premiership, the most number of goals scored at any one ten minute is not within the last ten minute period like Division One or the set rule that I have found. The Premiership has the period within 11-20 minutes at which the most goals are scored. Due to this I can see that the Premiership brings its largest number of goals in the 11-20 minute period rather than the last ten minute period as in Division One. As I stated before, this pilot study is not sufficient enough for me to base anything on, and so I will just keep its analysis and results in mind, but I will not draw any firm prediction from this. I will hopefully find what I am looking for in my second pilot study.