If my distribution is to be determined as normal, then 95% of the data will lie within 2 standard deviations of the mean and 99.8% of the data will lie within 3.
I will compare the reaction times of students to test my second hypothesis. I will draw comparative box plots and look for outliers. An outlier is any value which is more than 1.5 times the interquartile range below the lower quartile or 1.5 times the interquartile range above the upper quartile.
I will calculate averages of central tendency and spread, including the mean and standard deviation. I expect that reaction times will be faster and more consistent with the dominant hand. I think this because the muscles of the dominant hand are quicker, this is because the pathways are open more due to it being use more. The reaction times are faster because it is worked more.
I will calculate the variance and the standard deviation for my set of data to prove that this is the case.
As I am measuring time my data is continuous. Therefore I must use grouped data. This will affect my analysis. For grouped data I can only state the class interval that contains the median. I will draw a cumulative frequency curve and calculate quartiles.
To test my third hypothesis I will draw scatter graphs to see if there is a correlation between the speed in reaction time between one hand and the other. I will calculate the mean point and use this to draw a line of best fit and regression line. "Regression" is the process of finding the function satisfied by the points on the scatter diagram. I know that the points might not fit the function exactly but my aim is to get as close as possible. "Linear" means that the function we are looking for is a straight line (so our function f will be of the form y = ax + b for constants a and b).
I expect the two to be positively correlated, meaning that if one person records a fast reaction time with their dominant hand, then they will be fast with the other hand, albeit not as fast, as I hope to prove from earlier tests. I will calculate Spearman’s’ Rank Correlation Coefficient (SRCC) to see how strong the correlation is. I expect it to be towards a value of 1. I will also calculate the regression line of my scatter graphs
The main factors which will need to be considered whilst doing the experiment are:
- The sex of the student. Whether they are a girl or boy will affect their rate of reaction so I will be using the same amount of boys and girls.
- There sight will affect the rate of reaction for example they can see the ruler they will be more likely to catch it.
- The distance between their hand and ruler will affect the rate of reaction as they will be able to catch it quicker if it was closer to ensure its affair experiment I will be measuring 10cm above the hand each time I do the experiment so the distance will be kept constant.
- Whether they use their left or right hand will affect their rate of reaction as if there usually left handed it will be more difficult for them to catch it with their right hand. This is a factor I will be testing.
- The hearing of a student can affect the rate of reaction because if they were to hear someone shouts before the ruler was dropped they would prepare themselves for it.
- The age of the person will also affect their reaction rate as a younger person’s reflexes tend to be quicker.
- The health of the person will also affect their reaction rate as if there in good health they will be quicker. Everyone I am experimenting is in decent health.
- Also weather they had practice or nor because in theory practice makes perfect in this case your reflexes would quicker because they’d be prepared.
Prior to this experiment I did an experiment to help with a prediction and also to help indicate a number and range for the main experiment. The results from this experiment had many errors occur due to humanly errors. Whilst doing the experiment we didn’t measure the 10cm above the hand accurately. This caused all are results to be inaccurate also we decided we needed to be more careful when measuring the 10cm above people hand because they kept moving their hand as the ruler went up.
Also we only when we did our preliminary experiment we only did the experiment 5times. This didn’t give us a good range of results and caused the whole experiment to weaken. We noticed no adequate conclusion could be backed up with sufficient evidence as a suitable graph could not be formed.
We also tried different measurements from the hand and came to the conclusion 10cm would be adequate. From the preliminary experiments we decided to do the experiment 10times so we could monitor the investigation closely. So in our real experiment we decided on 10cm from the hand and we would repeat it 10times.
My data is primary data as I have collected it myself, I will use a stratified sample of 60 students from a range of year groups. Stratified sampling is when you divide your sample into different categories e.g. age or gender and then take a sample from each which is proportionate to the number in the overall population.