You are on page 1of 11

BY

Assignment 3 Manisha Mishra

2303
1) OBJECTIVE- TO SEE A SIGNIFICANT DIFFERENCE IN PERFORMANCE BETWEEN BOYS AND GIRLS

H0- There is no significant difference in the performance between Boys & Girls

Ha- There is a significant difference in the performance between Boys & Girls

If Performance variable HAS A NORMAL DISTRIBUTION

Then t test would be followed

>dataghs$Gender<-factor(dataghs$Gender,level=c(0,1),labels=c("Boy","Girl"))

> str(dataghs)

> t.test(dataghs$performance~dataghs$Gender,var.equal=T)

Two Sample t-test

data: dataghs$performance by dataghs$Gender

t = -0.31332, df = 198, p-value = 0.7544

tcritical = 1.972

Tcalculated<tcritical so null hypothesis is accepted

INTERPRETATION – Since Null hypothesis is accepted therefore there is no significant difference in


the performance between Boys & Girls
IF PERFORMANCE VARIABLE HAS NORMAL DISTRIBUTION

IF PERFORMANCE VARIABLE DOENOT HAVE A NORMAL DISTRIBUTION

>wilcox.test(dataghs$performance~dataghs$Gender)
data: dataghs$performance by dataghs$Gender

W = 4866.5, p-value = 0.8205

> count<-table(dataghs$Gender)

> count

Boy Girl

91 109

> barplot(count,main="Distribution of Gender",xlab="Gender")

Since p value >0.05 , H0 is accepted

Interpretation – Null hypothesis is accepted which means there is no significant difference in


performance between Boys & Girls
2) OBJECTIVE- TO SEE A SIGNIFICANT DIFFERENCE IN THE PERFORMANCE OF STUDENT BASED ON
HIS SCHOOLING

H0- There is no significant difference in the performance of student based on his schooling

Ha- There is a significant difference in the performance of student based on his schooling

If Performance variable HAS A NORMAL DISTRIBUTION

Then t test would be followed

>t.test(dataghs$performance~dataghs$schtyp,var.equal=T)

Two Sample t-test

data: dataghs$performance by dataghs$schtyp

t = -1.6261, df = 198, p-value = 0.1055 tcritical = 1.972

tcalculated< tcritical so null hypothesis is accepted

INTERPRETATION -Null hypothesis is accepted which means there is no significant


difference in performance based on the school type
count<-table(dataghs$schtyp)

> count

Public Private

168 32

> barplot(count,main="Difference in school type",xlab="schtyp")

IF IT DOES NOT HAVE A NORMAL DISTRIBUTION

>wilcox.test(dataghs$performance~dataghs$schtyp)

W = 2246.5, p-value = 0.1416

Since p value >0.05 , H0 is accepted

Interpretation – Null hypothesis is accepted which means there is no significant difference in


performance based on the school type
3) OBJECTIVE- TO SEE A SIGNIFICANT DIFFERENCE IN AVERAGE BETWEEN BOYS AND GIRLS

H0- There is no significant difference in the average between Boys & Girls

Ha- There is a significant difference in the average between Boys & Girls

If average variable HAS A NORMAL DISTRIBUTION

>t.test(dataghs$average~dataghs$Gender,var.equal=T)

Two Sample t-test

t = -0.31332, df = 198, p-value = 0.7544 tcritical 1.972

tcalculated< tcritical so null hypothesis is accepted

INTERPRETATION - Null hypothesis is accepted which means there is no significant


difference in average based on the gender

IF IT DOES NOT HAVE A NORMAL DISTRIBUTION

>wilcox.test(dataghs$average~dataghs$Gender)

W = 4866.5, p-value = 0.8205

Since p value >0.05 , H0 is accepted

Interpretation – Null hypothesis is accepted which means there is no significant difference in


average between girls & boys
4) OBJECTIVE- TO SEE A SIGNIFICANT DIFFERENCE IN AVERAGE BASED ON THE SCHOOLING

H0- There is no significant difference in the average based on the schooling

Ha- There is a significant difference in the average based on the schooling

If average variable HAS A NORMAL DISTRIBUTION

>t.test(dataghs$average~dataghs$schtyp,var.equal=T)

Two Sample t-test

t = -1.6261, df = 198, p-value = 0.1055

tcalculated< tcritical so null hypothesis is accepted

INTERPRETATION - Null hypothesis is accepted which means there is no significant


difference in average based on the schooling
> count<-table(dataghs$schtyp)

> count

Public Private

168 32

> barplot(count,main='Difference in school",xlab="schtyp")

IF IT DOES NOT HAVE A NORMAL DISTRIBUTION

>wilcox.test(dataghs$average~dataghs$schtyp)

W = 2246.5, p-value = 0.1416

Since p value >0.05 , H0 is accepted

Interpretation – Null hypothesis is accepted which means there is no significant difference in


average based on the schooling

You might also like