The Average Student Plan Essays

The Average Student Plan Essays The Average Student Plan Essay The Average Student Plan Essay Having looked at various other methods of data collection I have come to the conclusion that I will base my work on the average student. Firstly I think it is appropriate that I define what the average student actually is. There are different interpretations as to what an average student is, for adults perhaps teachers they may base their ideas as to what the average student is on exam results. Other people may believe the average student is based on what sports they play, what genre of music they listen to or what type of books they read. However for the data collection I will be carrying out I will base my data on a students physical attributes, as this will enable me to use higher-level maths techniques, which will be later explained. Due to limitations in working time I have decided to focus on recording just the height and weight of males and females. I have decided to focus on height and weight because they are continuous data therefore it will be possible to apply higher-level maths techniques. Another reason for selecting this data collection is that the data is easily accessible. There are two hundred and seven pupils in my year and obtaining their measurements will not be a problem. I will hopefully have the opportunity to compare my results with the National Data from the Child Growth Foundation (1996), which my teacher has recommended we find and use because using this I will be able to compare upper and lower quartiles and the medians. Firstly my aim is to research and discover some hypotheses about an average student using statistical techniques. The three hypotheses I have selected will be later explained For my coursework I have the opportunity to select from four methods of sampling. Not all sampling methods will fit with my data collection as some may affect the reliability of the results. The term given to this is bias meaning if an item in a population does not have an equal selection opportunity from a sample, it is said to be bias. An example of this that I am aware of and may occur in my collection is, from a population of forty there would have to be twenty males and twenty females for it to be fair. If it were any different to this then it would be bias. Here are the four methods of sampling I can choose from when carrying out my data collection. Systematic Sampling An example of this sample would be the selection of a 10% sample by going through the population picking every tenth item or individual. The disadvantage of this method is this would only provide a representative sample if the population was arranged in a random way and not in a way that might introduce bias. Information taken from OCR Graduated Assessment Stages 910 Textbook. Attribute Sampling The selection of the sample is made by choosing attribute such as head size and height from a list of people on the basis of their birthday being the first of the month and trying to identify any relationship between the two. Stratified or quota sampling The population is divided into strata or subgroups and the sample chosen to reflect the properties of these subgroups. An example of this would be if a population contained three times as many people under the age of 25 as over 25 then the sample should also contain three times as many people under 25. Information taken from OCR Graduated Assessment stages 910 Textbook. Random Sampling This is when there is no knowledge about a populations characteristics, for example any knowledge about the ages and gender in the population. In this case a sample has to be selected on the basis that all items are equally likely to be chosen. To ensure the sample is random and as accurate as possible the sampling must be repeated several times and then have the results averaged. The sample I have decided to use is systematic and random sampling this is because it would save a lot of time because my population is of two hundred people. In addition to this I have decided to work out an average using a percentage of the population. I have decided to use twenty percent, forty people, twenty males and twenty females to ensure it is a fair test. The hypotheses I will be investigating will be, * You would assume that as heights increase so to will weights. * You would think that on average boys are heavier than girls. * You would think that on average boys are taller than girls. Having these hypotheses I will be able to test the relationship between height and weight and also to compare the results for males and females. When comparing genders I will be using cumulative frequency, histograms, standard deviation and normal distribution. Once I have obtained my results I will transfer them to a spreadsheet so I can clearly set my results out. I will select the pupils I need for my investigation, twenty males and twenty females and measure their height and weight. I considered working with another person to help me but I believe I can work just as well on my own and still be as accurate. Each pupil in the sample will be treated the same this is to avoid an unfair investigation. When measuring height and weight each person in the sample will remove their shoes as this increases height and weight. I am aware that there will be complications through the investigation for example I need to check equipment in case they are slightly inaccurate. Perhaps other mistakes such as the student moving slightly will alter the readings and therefore not providing true results. I intend to be as accurate as possible and consistent with my recordings treating each student the same. Once my data has been recorded it then has to be sorted in order, firstly I will separate the data into genders and by using spreadsheets on the computer I can arrange the data into ascending order therefore when I use Cumulative Frequency Graphs I will be able to find the range numbers easily. I will be able to compare the different sets of results and from this point highlight any potential anomalies. With the results I obtain I have the opportunity to compare my results to the National Data. The appropriate average differentiates with graph work for example with Cumulative Frequency I will be working with median however I can compare this value with the mean data and comment on the graph distributions. Once I have the collected data I will then look back to my hypotheses and determine whether or not I was correct. As I mentioned earlier I will attempt to use higher-level maths techniques. I will be using Pearsons Product Movement Correlation Coefficient to test for any relationships between data. I will also include evidence of Cumulative Frequency diagrams and Box Plots as an attempt to prove any relationships between sets of data. I will also be using Histograms and Standard Deviation to find certain things such as the spread about the mean and using these values to compare with each set of results.