STATISTICS

Statistics



Statistics

Question A-1

Ordinal data scale should be used in this data due to reason that data is qualitative, in ratio scale data used is quantitaive.

Question A-2

The experimental design that should be used in this data set should be General linear model, reason is that both variables are qualitative therefore they should be analyzed through GLM technique. This technique is selected instead of simple linear regression due to nature of data set.

Question B-1

The researcher should go fro discriminent analysis in this data set becase variables are catogorical.

Question B-2

Hypothesis 1: there is the significant impact of Gender on body weight

Hypothesis 2: There is the significant impact of Drink given to child on their body weights.

Question C-1

In simple linear regression, we forecast tallies on one variable from tallies on the second variable. The variable we are basing our predictions on is called predictor variable and is mentioned to as X.

Question C-2

Residual contrive is the graph that exhibitions residuals on upright axis and unaligned variable on level axis. If points in the residual contrive are randomly dispersed around level axis, the linear regression pattern is befitting for details and figures; additional, the non-linear pattern is more appropriate.

The residual contrive displays the non-random convention - at odds residuals on decreased end of X axis and affirmative residuals on high end.

Question C-3

John

Betty

Sarah

Peter

Fiona

Charlie

Tim

Gerry

Martine

Rachel

English score

78

70

81

31

55

29

74

64

47

53

Time

12

32

19

31

30

15

22

10

17

16

In this case, scatter gram shows no particular pattern. It is clear that we can't draw the straight line anywhere near facts and figures points, and we state that there is no association between extent of time taken to journey to college and last English assess that the scholar gets. We cannot forecast English assess of any scholar based on how long it takes him to get to college.

Question C-4

Correlation is what you are doing when you contrast two groups of measurements (each set is called the variable).&nabs; If you were to assess everyone's size and weight, you could then contrast heights and weights and glimpse if they have any connection to each other -- any "co-relation," if you will. 

Let's take a demonstration in this regard. A perfect correlation is +1.  For example, here is some data

Shoe size

Foot length(inches)

John

4 1/2

9 1/4

Dave

5

9 3/8

Sam

5

9 1/4

Jim

6 1/2

9 1/2

Ed

6 1/2

9 3/4

Bob

7

9 3/4

Ted

8

10 1/8

Matt

11 1/2

11

Damian

12

11 1/4

Horton

14

11 3/8

We can organize facts and figures on the journal like this:

This is called the scatter plot; The line is line that recounts "best fit" -- in other words, ...