I have produced a Composite Bar Chart which is attached to the upcoming page
The Pie Chart shows the percentage of each year in the School and these help me form my sample and this form of representing data is efficient accurate and eye catching and help form divisions in my data.
Calculations:
Year 7: 282 Divided by 1183 Multiplied by 360
= 86 Degrees
Year 8: 270 Divided by 1183 Multiplied by 360
= 82 Degrees
Year 9: 261 Divided by 1183 Multiplied by 360
= 79 Degrees
Year 10: 200 Divided by 1183 Multiplied by 360
= 61 Degrees
Year 11: 170 Divided by 1183 Multiplied by 360
= 52 Degrees
Total 360 Degrees
Mean and Mode of Frequency Data
I will now find the Mean, Median and Mode of the Frequency that I have found and this will be quick efficient and reliable and will help me gain evidence on whether boys are taller and weigh more in comparison to girls.
Mean of Girls and Boys Weight
Mean = Total Frequency Times Midpoint Divided by Total Frequency
980 Divided by 20 = 49 kg for BOYS
Mean = 895 divided by 20 = 44.75 rounded off to 45 kg for GIRLS
Mean of Boys and Girls Height
Mean= 3250 Divided by 20 = 162.5 = 163 cm = 1.63m
Mean: 3130 Divided by 20 = 156.5 which is 157 cm = 1.57m
Comparison of Mean of Girls & Boys (Height and Weight)
In comparison of the mean I have found for both boys and girls heights the differences vary. This shows that on average from my sample boys are weigh more then the girls in my data although it may not be by a large amount. Boys from my Stratified Sample also on average are 6 cm taller than Girls.
Mode of Girls and Boys Weight
The arithmetic mean of a group of numbers is found by dividing their sum by the number of members in the group; e.g., the sum of the seven numbers 4, 5, 6, 9, 13, 14, and 19 is 70 so their mean is 70 divided by 7, or 10. Less often used is the geometric mean (for two quantities, the square root of their product; for n quantities, the nth root of their product).
Modal Weight Stem and Leaf Diagram
Key 3 / 4 = 34 kg
To find the modal weight I will now look at which frequency seems to have appeared the most often and for the Boys Modal Weight it is:
Modal Group for Boys: 40 ≤w< 50 Modal Weight for Girls: 40 kg
Below is Similar Data but Different Calculation methods to find the Mean, Median and also the Mode as this will help me towards proving my hypothesis also this data will help me find the spread and the average of the data which will be helpful throughout this portfolio:
Boys: Height
Mean
1.38 + 1.61 + 1.63 + 1.50 + 1.53 + 1.82 + 1.62 + 1.38 + 1.63 + 1.53 + 1.61 + 1.52 + 1.32 + 1.56 + 1.80 + 1.68 + 1.70 + 1.62 + 1.65 + 1.8
20
= 31.28 Divided by 20 = 1.6m
Median: 1.65 m (Calculated with the Use of Microsoft Excel)
Range: 1.8 – 1.32 = 0.48 m
Boys Weight
Mean: 48.1 kg
Median: 48 kg
Range: 64-32 =32 kg
Girls: Weight Girls Height
Mean Mean
30 + 36 + 36 +37 + 38 + 40 + 40 + 40 + 41 1.56 m
+ 45 + 46 + 47 + 47 + 52 + 52 + 52 +56
57 + 59 + 60
20
= 45.55 which is rounded to 46 kg
Median: 46kg Median 1.56m
Range: 60 – 30 = 30 Range 1.75 – 1.41 = 0.34m
Histogram and Frequency Polygons
From the data I have collected and formed through my frequency tables and mean averages and many more I will now produce a Frequency Polygon and a Histogram that shows the Boys & Girls Height and Weights From my sample that I have taken a for my assignment. The Frequency Polygon will clearly identify the shape of my variations and both these forms of representing data will help me form a sufficient analysis.
Boys Height
Histogram for Boys Weights
Girls Height and Weight Frequency Polygons & Histogram
These histograms now give me a clear picture of the data distribution. For the sample there is an even distribution of data. The middle group has the highest frequency which is expected. For the data to be evenly distributed, the other two sides must be fairly symmetrical. It is clear that the histogram do not show this. This shows that the majority of scores were above the median.
Girls Height
These Representations of Data shows admirably that the average height is 150 to 160 which I believe is slightly above average in my honest opinion for my sample and also I have come to find that the trend is rather varied although the frequency are upward to a certain point and downward from the peak onwards.
Girls Weight
This data shows with great intent that the highest frequency I 40 – 49 kg which shows that both boys and girls from the stratified sample that I have taken for this area of my assignment and this hypothesis in particular that they have a lot In common in terms of frequent data sources. In addition to this I have also come to find that the girls do not have many students above 69 kg whereas for boys there are 3 as times as many students above this height.
Cumulative Frequency Diagrams & Tables
I will now produce cumulative frequency diagram for both girls & boys height and weight and this will help me gain sufficient evidence towards forming my conclusion and I will also find percentiles and will produce box and whisker plots as this will help me view my data and trend efficiently.
1) BOYS HEIGHTS:
Using a Statistical Program that I have downloaded I entered my data into the appropriate field and below Is a diagram of for the Cumulative Frequency which will help me identify percentiles and trends in my data. The graph will also include a line which helps me identify the trend clearly
2) BOYS WEIGHT
From this particular Cumulative Frequency Diagram I have been able to find that the average weight from the sample that I have taken is 60kg which I my opinion is generally quite high and the Interquartile Range that I will be finding at a later stage will help me view the spread of data and the margin of error. I believe that all the graphs that I will produce will help me complete my objective and conclude efficiently and successfully.
Boys Height Box and Whisker Plot
I then used the same statistical software I created a box and whisker which is a vital representation of data and it will
Boys Weight Box and Whisker Plot
Comparison of Box and Whisker Plots
Form this Box and Whisker Plot I will be able to find the median which will show the middle frequency of my data and also will be able to view the maximum and minimum values for both height and the weight and find the percentiles and the quartiles. For the Height the Median is 170cm and the interquartile range is 30 cm and the maximum value is 200 cm and minimum value is 140cm. whereas for the weight I have found the Median as 60kg which is respectively what I had predictable and is suitably accurate although the range of data for the weight is less and the data is negatively skewed in comparison to the height for the boys.
I will no be producing Cumulative Frequency Diagram and Tables for the Girls Height and Weight since this will help me form a sufficient and reliable comparison in height between Girls and Boys and form a successful and accurate conclusion to my hypothesis.
GIRLS HEIGHT
Cumulative Frequency Diagram
GIRLS WEIGHT
Cumulative Frequency Diagram
My variable will be the Height
Median 60 kg
Lower Quartile 50 kg
Interquartile Range 22g
Upper Quartile 72 kg
From this I have found the spread on data efficient and on the following page I will compare my results from the cumulative frequency against both boys and girls height and weight and make a suitable conclusion from this representation.
Comparison of Height and Weight of Boys and Girls from C.Frequency and Box Plots
From the Cumulative Frequency Diagram I have come to find that the Median height for boys in the sample that I have taken is larger in comparison to the girl’s median height. Boys Median Height is 170 cm whereas Girls Median Height is 162 cm and shows that there is an 8cm difference in the middle figure from both sets of data I have collected although this may on some occasion be accurate due to the fact my sample may not be efficient and also the range of data varies where the IQ range for Boys is 21 cm and Girls in 33 cm which shows that the spread of data that I have from my sample for boys is narrower whereas girls have a wider spread and make the results reliable as a whole. As far as the Weights are concerned the range is similar and the range is rather symmetrical also and this may be since my graphs may be irregular.
Scatter Diagram
I will now produce a Scatter Graph which shows the Height vs Weight for all the Boys and girls data that I have collected and I will be able to find whether there is a pure correlation and I will then compare my results and I will see whether my two sets of data are related from my sample:
I will conclude after both graphs
Boys Height vs Weight
Girls Height vs. Weight
Conclusion of Scatter Graphs
Boys & Girls Height vs. Weight = From the Scatter Graph that I have produced I have come to find that there is a Correlation or a Trend between both variables which are Height and Weight and there is a Fairly Strong Positive Correlation and this shows me that the taller the person the higher the weight although there are always some irregularities which is unique as a symmetrical trend is rather impossible as every human being has various growth period. In addition to this will be the major form of representation in my and I believe this shows the trend clearly and efficiently.
Body Mass Index
Calculation= WEIGHT / (HEIGHT) ²
I will be Doing this calculation for 5 people randomly from the sample that I have taken:
Calculation= WEIGHT / (HEIGHT) ²
-
43/(1.54) ² = 43/2.3716 = 18.13
-
56/(1.61) ² = 56/2.5921 = 20.64
3) 37/(1.49) ²= 37/2.2201 = 16.67
4) 56/(1.72) ²= 56/2.9584 = 18.93
5) 60/(1.75) ²= 60/3.0625 = 19.6
From this calculations I can conclude that the Higher the Height in Equivalent to the Weight the Higher the Value of the Body Mass Index which shows me that both variable have a certain trend and also shows that on some occasions the height is parallel to the Weight.
Spearman Rank Correlation Co-Efficient
I will now using the same data sample of five random students to find the spearman rank correlation. The Spearman's Rank Correlation Coefficient is used to discover the strength of a link between two sets of data:
R = 1 – 6(2) = 1- 12 = 1 – 12/120 = 1 -1/10 = 0.9
5(25-1) 24 Times 5= 120
Evaluation of Hypothesis 1
In my honest opinion I feel that I successfully completed and analyzed my hypothesis and I have gained a sufficient evidence to back up my theories. I would like to remind you that my main objective for this hypothesis was to find out whether I was correct or incorrect in my thinking that Boys at Mayfield School are taller and weigh more on average than the Girls at the same school. Within this aim I was also aiming to find whether there is a certain trend or relationship between the height and weight of the students that I have chosen to analyse and as I explained earlier due to the large number of students I was not possible to analyse all students so I gained a sufficient sample which I made as unbiased as possible. MY HYPOTHESIS WAS CORRECT
Conclusion of Hypothesis
- The Histograms, frequency polygons proved that the results were more accurate and made more sense than that from the random sampling.
- There is a positive correlation between height and weight. In general tall people will weigh more than smaller people.
- In general boys tend to weigh more and be taller then girls.
- By doing stratified sampling, there were a fewer exceptional values caused by different year groups and therefore ages. I was bound to find irregularities within my data
- The cumulative frequency curves confirm that boys have a more spread out range in weight, with more girls having smaller weights. In height, boys tend to be taller.
- The spearman rank correlation coefficient shows that the correlation between height and weight is strong.
- My Body Mass Index showed that there is a strong trend between height and weight
- In general the taller a person is, the more they will weigh.
- There is a positive correlation between height and weight. In general tall people will weigh more than smaller people.
- There therefore is a positive correlation between height and weight across the school as a whole. This correlation seems to be stronger when separate genders are considered
- If I had taken larger samples my hypothesis may become more accurate.
Hypothesis 2 ‘Key Stage 4 Students who watch more hours of television on average have a lower IQ Level’
Planning
For this particular hypothesis I will be only be using the data of the School Year of 10 and 11 due to the fact I have specifically chosen to investigate Key Stage 4 and this is the Stage studied by these 2 school years. I have chosen to only base my investigation on this key stage since I believe if I used the complete Mayfield High School data for this hypothesis the range and spread and range of data would be too large to make a sufficient analysis of the results that I will gain. I will be a Sampling Method is which I can sufficiently break down the number of student and meanwhile keep the investigation fair as possible.
Sampling
The sampling method that I have chosen to use for this very particular hypothesis is RANDOM SAMPLING. This is where every item in the population will have an equal chance of being selected. Below is the method that I chose to do the successfully and conveniently. I believe this method will help my results and outcome stay unbiased.
- I Printed a Copy of the Students and there personal qualities and detail.
- I then cut out the student First Name and Surnames.
- As you may remember from my two ways table earlier in this portfolio the total number of student that I have in this particular investigation is 370.
- In my sampling the student Gender is not an issue of comparison so I will be using all the names in one sample
- I then used a Hat which I had and put the all the names into this Hat
- I then shook the hat so that I cannot tell or neither can anyone else see the order of the names, which are in the hat.
- I then instantly decided to withdraw 30 students from the hat. An unbiased person who was besides me while I was completing this sample adjudged this procedure.
After I had sufficiently completed the sample below are the results that I had obtained. I then typed the results into a Microsoft Excel Spreadsheet. Below is the data that I have gained from my sample. I have decided to only include the Name, Year, Number of Hours of TV, and Favourite TV Show and IQ Level since these is the field, which are important to me whereas other fields are invaluable.
Random Sample Results:
I will now begin to investigate the hypothesis that I have chosen to the fullest extent possible I will first begin by producing frequency tables to firstly find the mean and the average IQ Level of the Data that I have chosen and also the average number of hours of TV watched per student.
Mean IQ of Student in my sample
I will now find the Mean, Mode of the Frequency that I have found and this will be quick efficient and reliable and will help me gain evidence on the average IQ and see whether each student has a IQ high or low in comparison to the mean of the sample.
Mean can be calculated using this formulae shown below:
Σfx Divided by Σf = 2840 Divided by 30 = 95.7 1d.p
Average IQ Level is 96.
From this excellent presentation of data I have come to find that 7 Students out of the 20 students are below average that is rather a concerning since this is a large sum of students. Also in addition to this I will be able to now compare these student hours of TV and this will help me towards forming a conclusion on the trend between both the number of hours of TV and the IQ level and prove whether my theory was correct or incorrect. I can also gain from this that the most IQ group or median group is of 100≤w<110 and this group consists of over 60 percent of the student that I have sampled.
Mean TV per week of Student in my sample
As with the first mean frequency table that I had produced this will help me toward making a comparison and to realize a trend that there may be between the number of hours of TV watched per weeks and the particular students IQ this is my aim from producing the Mean Frequency Tables
Mean can be calculated using this formula shown below:
Σfx Divided by Σf = 540 Divided by 30 = 18 Hours
Average Hours of TV watched by a student in my sample is 18 Hours
From this particular presentation of data I have come to find that over 50 percent of Students watch more TV than the average that I have found of my sample after producing a mean frequency table.
Comparison
I can compare that from the tables that I have produced and also by looking straightforwardly at the sample I have created which is that the children who watch a large sum or amount of TV on average per week rather peculiarly have an IQ which is above average in contrast to the average IQ I had found for my sample. Although I have also found using my general knowledge and understanding and interpretation of the data that the students who watch a reasonable amount of TV have a good IQ level also so this shows me that there is a certain limit to the hours of TV to be watched. Contrary to this TV can help the mind be stimulated and also help to take immediate action in everyday life situation and make you more aware of what is occurring in your surrounding and environment whatever program it may be.
Mode of Girls and Boys Weight
The arithmetic mean of a group of numbers is found by dividing their sum by the number of members in the group; e.g., the sum of the seven numbers 4, 5, 6, 9, 13, 14, and 19 is 70 so their mean is 70 divided by 7, or 10. Less often used is the geometric mean (for two quantities, the square root of their product; for n quantities, the nth root of their product).
Modal Weight Stem and Leaf Diagram
Key 13 / 0 = 130 kg
Histogram
This form of data representation will help me find the spread of data and also the trend in data and the results and firstly below is a histogram showing the IQ levels of my sample:
IQ
Average Hours of TV per week
This shows me the Median Group for IQ Level 100-110 and the Number of Average hours varies Median group of 10-30
Scatter Graph
I will now produce a Scatter Graph. I will use this to compare the two sets of data when there are two variables, which are IQ/Hours of TV. Also in addition to this a straight line is used to join the dots and this will help me view any trend between the two variables that there may be. The Data I will input in Microsoft Excel and this will help me produce an efficient and successful graph in which I can then evaluate on the trend.
LINE OF BEST FIT TO SHOW TREND CLEARLY IS ALSO INCLUDED.
This Scatter Graph that I have produced shows that there is a very weak positive correlation between the two variables that I am investigating for this hypothesis and this means that that the increase in Hours of TV the higher the IQ on average which is rather bemusing as I had expected the opposite using my general knowledge and everyday theory although the statistics and my sample have proven me wrong at this current moment. It can also be easily viewed that although I may be incorrect there are some irregularities in the sample or variances in that particular person characteristics and personality. Also I feel this trend may be since the TV helps stimulate a child mind to a certain degree and whichever program it may be it has some effect on the way a child lives there live and in my sample the effect is largely positive as I have duly come to find. The variances also show that my sample is not unbiased and reliable as it is natural and unique for they’re to be some minor irregularities. Overall there is a casual relationship between the IQ Level and Hours of TV watched for Children at Mayfield School
Variance and Standard Deviation
Variance is a measure of the spread of the distribution and the square root of the result in the Standard Deviation. This can be calculation using a certain formula, which is shown below. By completing this statistical task it will help me measure the spread of the data and the mean of the distribution.
Standard Deviation Formula:
Below I will now produce a table in which I will calculate the standard deviation for firstly the IQ and then the Average Hours of TV:
IQ Levels
The Mean Is
(3 X 85) + (5 X 95) + (19 X 105) + (1 X 115) + (125 X 2)
= 3090/30= 103
30
And the standard deviation will be
√ [2 X (85-103)] + [5 X (95 – 103)] + [19 X (105-103)] + [1 X (115-103)]
+ [2 X (125-103)
30
√ 18/30 = 0.7746 4.dp
I will now complete the same Standard Deviation Calculation as I had produced but in this particular case I will do the Deviation and Spread of Data on Average Hours per Week per student
Average Hours Per week TV
The Mean Is
(5 X 7) + (10 X 15) + (10 X 25) + (3 X 35)
= 540/30= 18
30
And the standard deviation will be
√ [7 X (5-18)] + [10 X (15 – 18)] + [10 X (25-18)] + [3 X (35-18)]
30
√ 1/30 = 0.0333 4.dp
Conclusion
From the Standard Deviation that I have come to find how spread out the data that I have used in my sample for this particular hypothesis and measure of the dispersion of the data and has helped me see whether the data I have is unbiased and of a sufficient spread and size and it will also help me measure the reliability of the sample that I have taken is that it is reliable which is usual on a scale between –1 to 1 and also that the spread of data in the IQ levels in larger than the Average TV hours watched which shows me that in my final evaluation the results for Average TV hours watched per week will be generally more accurate and conclusive.
Standardised Scores
I will now find the Standardised score for 5 students from my sample at Random. It will help me compare the values.
5 Student that I have chosen for this calculation are:
Standardised Scores Calculation/Formula
Score – Mean
Standard Deviation
Elise Standardised Scores
IQ Level = (101-103) / 0.7746 = - 2.58
Average TV Hours = (10-18)/0.033 = -240
Katie Standardised Scores
IQ Level = (104-103) / 0.7746 = 1.29
Average TV Hours = (17-18)/0.033 = -30.30
Justin Standardised Scores
IQ Level = (100-103) / 0.7746 = - 3.87
Average TV Hours = (24-18)/0.033 = 181
From this I find that that Katie has the better standardized result since her results is a positive figure whereas as Justin’s standardized score was a negative value and Elise has the best Average TV house standardized score.
Time Series Line Graph
A Line graph is used to display data when the two variables are not related by an equation and you are not certain what happens from one point to another. I will need to plot this graph extremely carefully due to the fact I cannot enter all the data in my sample and produce this graph since there will no trend which I will be able to view so I will viewing student at different Average Heights in Order and then Plot their IQ Levels on a Line Graph and from this I will able to make a trend comparison. The Line graph will consist of TWO Variables.
On the Whole it is very compact and difficult to give a definite conclusion according to this graph due to the fact the trend is fluctuation up and down and very slightly and upward trend although it is very difficult to tell and verify and form an definite answer to the hypothesis that I have chosen to investigate. Also in addition to this the verification are rather minor and steady and fluctuate largely. Also I can add to this that the basic IQ level is not efficient and excessively effected by the Number of Hours of TV you but generally your age and your understanding and knowledge overall.
*I WILL BE CONCLUDING AND EVALUATING THIS PARTICULAR HYPOTHESIS AFTER THE COMPLETION OF THE THIRD HYPOTHESIS
Hypothesis 3
‘ Left Handed Students have higher IQ levels and Key Stage 2 Results in Comparison right handed student at Mayfield High School’
Planning
For my third and final hypothesis I will need the data of Mayfield High School between the years of 7 to 11 and this is due to the fact I need a wide sampling frame so that my analysis and results are not unbiased and as accurate and efficient as possible. Also in order to make my portfolio unique I will be forming a SAMPLE in which I can accurately and efficiently minimise the range of the data but meanwhile keep the hypothesis as unbiased as possible as I had explained at an earlier stage.
The Sampling Method I will be using to do this is STRATIFIED as with the first hypothesis since I duly believe that this sampling method is the most accurate and reliable and meets the needs and requirements of this particular hypothesis to the fullest extent. I will now explain its benefits and the method in which I calculated and completed the Sample.
I will be calculating my Stratified Sampling using the Table above now and I will need calculation proportion to stratify my data spread/range.
Here is another copy of the two-way table of the table, which helps me form my sampling method.
On the upcoming page you will find the calculation of my Stratified Sample.
Stratified Sampling Calculations
I want 20 Students from the school total 1183.
I have come to find using Microsoft Excel that 251 Students throughout the school are Left Handed whereas Right Handed Students are 922 and there are 10 Both Handed so these student have been left out of my investigation for this hypothesis. As I have found Right Handed Student is common for most student.
Right Handed Students= 922 Divided By 1173 multiplied by 20 = 16
Left Handed Students= 251 Divided by 1173 Multiplied by 20= 4
Below are the Results of my Sample
I will now begin my Statistical Investigation on this particular hypothesis and I will represent my data in many various methods and also produce analyse after each graph that I will be producing. I will begin with firstly forming a pie chart at the number of students in my Sample that are Right and Left Handed and also a multiple bar chart showing the number of students investigated.
Pie & Bar Charts
Probability
Probability is an area that is used to predict the chance of something happening in the future or the likelihood of something to occur successfully and efficiently.
Here is the likelihood of a student being Left Handed from my sample:
0 1
I will now analyse data further in depth the sample that I have collected and form a Two Way Table that shows the boys and girls that are Right & Left Handed:
Probability Tree Diagram
My variables are Male and Female and Left and Right Handed.
M= MALE R= RIGHT
F=FEMALE L= LEFT
R = RM
M
L = ML
R = FR
F
L = FL
Calculations have been written by Hand Below:
I have decided that if I am to compare my hypothesis efficient and to a certain high standard then I will have to analyse the data by looking at the data and picking put certain data sample and compare the results and form a hypothesis:
- FIRSTLY I WILL FIND TO SIMILAR STUDENT WHO ARE SIMILAR IN CHARACTERISITICS AND PERSONALITY
- BELOW IS THE DATA THAT I WILL BE USING:
- AS YOU CAN SEE FROM THE DATA THE DATA IS REASONABLY SIMILAR AND I WILL BE ABLE TO FORM A SUCCESFUL CONCLUSION FROM THIS.
*Conclusion for both Hypothesis 1 and Hypothesis 2 can be found on the upcoming page
Conclusion/Evaluation of Hypothesis 2
‘Key Stage 4 Students who watch more hours of TV on average have a Lower IQ Level.
I feel that I have successfully completed and investigated my hypothesis to an extent in which I can be sure of my accuracy of my conclusion and I gained many sufficient forms of evidence. As stated above my hypothesis was to find whether Students who spent a large sum of time per week watching TV have a lower IQ Level and I had come to find that my theory was incorrect to a certain level. I had used many statistical representations to prove my theory.
I was also aiming to find whether there was a regular trend between the two variable and aimed to make my hypothesis as unbiased as possible.The Histogram that I had produced showed that my results were unbiased and relatively accurate in comparison the stratified choice of methods and made more sense on the whole.There is a very weak positive correlation between IQ and TV hours although it is not enough to prove whether my theory was correct and there was not a strong positive correlation that I had expected. In general IQ Level does not increase an Hours of TV increase.Some irregularities were found. The Standardised Score and Standard Deviation help me find the spread of the data so the sample is unbiased and insufficient.
If I had taken large sample my hypothesis may become more accurate and able to form a successful conclusion. The IQ also depends on the persons surrounding, ability and knowledge and stimulation and motivation which can all play a factor in the results.Overall I have found that my Hypothesis was incorrect and the statistical evidence that I had gained did not back up my theory. Another reason behind my misfortune is the range of data from 7-11 is too wide and I should have narrowed the frame down but now helps me in the future.
Conclusion/Evaluation of Hypothesis 3
‘Left Handed Students have higher IQ levels and Key Stage 2 Results in
Comparison Right handed student at Mayfield High School’
In my honest opinion I feel that I have proved my hypothesis in a short span and I feel that the statistical evidence that I have duly gained is more then sufficient to form a hypothesis and if I used any other forms of representing data it would have meant I was generally repeating. Also I feel that I have used the correct statistical evidence to prove my theory and used the right evidence. In addition to that I feel that Left Handed students are generally have higher IQ Levels according to my sampling frame and investigate in this is what I have based my theory upon.