0% found this document useful (0 votes)
11 views22 pages

T-Test and ANOVA Analysis Guide

The document outlines various statistical tests including one-sample t-test, independent sample t-test, paired t-test, one-way ANOVA, regression analysis, and correlation analysis. Each section provides a procedure for conducting the tests using SPSS software, along with hypotheses, results, and inferences based on p-values. The document serves as a guide for analyzing data related to sportsmen's heights, students' aptitude marks, salesmen's performance, crop yields, and the impact of R&D expenditure on profits.

Uploaded by

326071029
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
11 views22 pages

T-Test and ANOVA Analysis Guide

The document outlines various statistical tests including one-sample t-test, independent sample t-test, paired t-test, one-way ANOVA, regression analysis, and correlation analysis. Each section provides a procedure for conducting the tests using SPSS software, along with hypotheses, results, and inferences based on p-values. The document serves as a guide for analyzing data related to sportsmen's heights, students' aptitude marks, salesmen's performance, crop yields, and the impact of R&D expenditure on profits.

Uploaded by

326071029
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

EXERCISE 6

T TEST ( ONE SAMPLE T - TEST)


The height of 12 sports men is 175 cm, 168 cm, 175cm, 170cm, 172 cm, 168 cm, 170cm, 175cm, 165cm,
16Scmand t72cm. Using one sample t-test compare the mean score of sportsmen's height.( test value
as t 70cm)
purpose: one sample t-test compares the mean score of a sample to a known value .The known value
is
usually the population mean.
Aim: To test the mean height of sportsmen.
Proceedure:
Get into SPSS window and open an untitled SPSS worksheet.
2. Click the variable view in the SPSS worksheet and enter variable name, type ofdata , width of the
data, decimal, value , label for the variable names values with possible options scales and
measurement .
3. Click data view, the data entered in variable view appears as column names. Then enter the data for
the column in the sequence until required.
4. Enter the data given and save the data.
S. Go to an.Äe and select compare means, choose t-one sample test .A dialog box of test variable
appears.
6. In the test value box enter the height of sporismeii as 170 cm.
7. Clicks continue and then OK.

Result: Thus the program is executed and verified successfully.


FLOW CHART
Hypothesis
HO: The is no significant difference between the mean height of sportsmen
HI: The is significant difference between the mean height of sportsmen

Flowchart

'ANALYZE

COMPARE MEANS

oy SAMPLE T-TEST

TEST VARIABLE

'TESTVARIABLE
HEIGHT OF SPORTS MEN

OK
OUT PUT

ONE SAMPLE STATISTICS

Menn Stnndnrd Standard Error


Deviation Menn
Ileights 12 170.667 3.2286

ONE SAMPLE TEST

Test value -

Sig (2 tailed) Mean confidence


difference interval of difference

Lower Upper

ileights .7t5 .489 -1.3847 2.7181

Inference :
Since p value 0.489 is greater than 0.05 we accept the null hypothesis and reject the alternate
Ilence there is no significant ditTerencebetween the mean heights of males.
EXERCISE
T TEST (INDEP DENT SAMPLE )
In an aptitude test the marks obtained by 2 different batch students in which the first batch
comprises of 9 students and the second batch has 11 students are given below.

Batch 1 65 78 75 60 65 59 78 69 68 o O
Batch 2 62 70 71 62 60 56 75 64 72 68 66

Examine whether the marks obtained by both the batch students are significantly different at
5% level
Aim: To examine whether the aptitude marks obtained by both the batch 1 a d batch 2 students are
significantly different at 5% level

Proceedure:
l. Get into SPSS window and open an untitled SPSS worksheet.
2. Click the variable view in the SPSS worksheet and enter variable name, type of data , width of
the data, decimal, value , label for the variable names values with possible options scales and
measurement .
3. Click data view, the data entered in variable view appears as column names. Then enter the
data for the column in the sequence until required.
4. Enter the data given and save the data.
5. Go to analyze and select compare means , choose Independent sample T- test .A dialog box of
test variable appears.
6. In the Independent sample T- test a dialog box opens, in which marks are entered in test
variables , group variables as students and defining group 1 as one and group 2 as two.
7. Click continue and them OK.
FLOW CHART
Jlypothcsis
110:The is no significant difference between the marks obtained by batch 1 and batch 2 in the
aptitude test
HI: The is significant difference between the marks obtained by batch 1 and batch 2 in the
aptitude test

Flowchart

'ÅNÅ11fYfE

NS

JNbéPEND , SAMPLE
EST/

jTßT-OAÅIÅBLE / i ROIjp i o;hYIÅBLE DEFINING ciRQÜbk


GROUPI
GROUP .
OUT PUT

ONE SAMPLE STATISTICS

Standard Standard
Students Mean
Deviation Error Mean

BATCH 1 9 68.556 7.16085 2.38695

BATCH 2 11 66.000 5.77927 1.74252

INDEPENDENT SAMPLE E TEST

Test Value = 170


MARKS f sign t Sig(2 Mean Standard 95% confidence
tailed) difference Error interval of
Difference difference
Equal Lower Upper
variances
assumed .440 .516 .884 18 .384 2.55556 2.89010 -3.51631 8.62742

Equal
variances .864 15.318 .401 2.95532 -3.7321 8.84328
not
assumed

Inference ;
Since p value( 0.516) is greater than 0.05 at 5 percent level of significance we accept the null
hypothesis and reject the alternate hypothesis. Hence there is no significant difference between
the aptitude marks obtained between batch 1 and batch 2 students.
EXERCISE/
PAIRED T TEST (INDEPENDENT SAMPLE )
A company arranged an intensive training on sales and marketing for its team of salesmen.
Following data shows a random sample of 10 sales men total sales made (in 1000's) in the weeks
immediatelybefore and the after the training .

Sales men 1 2 3 4 5 6 7 8 9 10

Before training 12 23 5 18 10 21 19 15 8 14
After training 18 22 15 21 13 22 17 19 12 16

Test whether there is an evidence of increase in mean sales achieved by sales men after attending the
training program.
Purpose: Paired t-tcst is a statistical technique that is used to compare two population means in the
case of two samples that are connected.
Aim: To test the significant difference between the sales achieved by sales men before and after
training
Proceedure:
1. Get into SPSS window and open an untitled SPSS worksheet.
2. Click the variable view in the SPSS worksheet and enter variable name, type of data , width of
the data, decimal, value , label for the variable names values with possible options scales and
measurement .
3. Click data view, the data entered in variable view appears as column names. Then enter the
data for the column in the sequence until required. Enter the data given and save the data.
4. Go to analyze and select compare means , choose paired T- test
5. In a paired t-test a dialog box appears, enter sales before as variable 1 and sales after as
variable 2.
6. Click continue and them OK.

Result: Thus the program is executed and verified successfully.


FLOW CHART
Hypothesis
HO: The is no significant difference between the sales achieved by sales men before and
after attending the training program
HI: The is significant difference betweenthe sales achieved by sales men before and
after attending the training program

Flowchart

ANALYZE

COMPARE MEANS/

PAIRE

VAI\1öBLE 21
PAIR VARIABLE 1
SÅLES AFTCR
SALES BEFORE! ! TiUMNINCj•
TRAINING

CONTINUE

0K
OUT PVT

PAIRED SAMPLE STATISTICS

Mean Standard Deviation StandardError


Mean
Paired
Sales Be&re 14.50 5.836 1.845

SalesAfter 17.50 10 3.567 1.128

PAIRED SAMPLE STATISTICS

Mean Standard Deviation Standard Error


Mean
Paired

Sales before training 14.50 10 5.836 1.845

Sales after Training 17.50 10 -3.567 1.128

Paired sam le correlation


Standard Deviation Standard Error Mean
Paired
Sales before training &
Sales after Training 10 .841 .002

Mean Standard Standard Lower Upper t df sig


Deviation Deviation
Error
Pair 1

Salesbefore
training & -3.000 3.432 1.085 -5.455 -.545 .2.%4 9 0.002
Salesafter
Trainin

Inference :
Since p value ( 0.002 ) is less than 0.05 at 5 percent level of significance we reject the null
hypothesis and accept the alternate hypothesis. Hence there is significant difference in means of
sales achieved by salesmenbefore and after the training program.
EXERCIS
F TEST (ONE WA ANOVA)
fields for the three variety of
The followingdata shows the yield(in tonnes) recorded in five different
crops sownin a year.

2 3 4 5
Crop/ field 1

20 21 23 16 20
Rice
18 20 17 15 25
Barley
25 28 22 28 32
Maize

sown in
Examine whether there is significant different in the yield of the 3 different variety of crops
the fields in a year .

crops
Aim: To Test whether there is significant different in the yield of the 3 different variety of
sown in a year
Proceedure:
1. Get into SPSS window and open an untitled SPSS worksheet.
2. Click the variable view in the SPSS worksheet and enter variable name, type of data , width of
the data, decimal, value , label for the variable names values with possible options scales and
measurement .
3. Click data view, the data entered in variable view appears as column names. Then enter the
data for the column in the sequence until required.
4. Go to analyze and select compare means and select one way ANOVA (F- test) from the
available options . Thus the given data is saved.
5. A dialog box opens in the window, type ' yield' in the dependent list , 'variety' in factor,
'options' as descriptive and 'post hoc' as Duncan .
6. Click continue and then OK.
FLOW CHART
Hypothesis
HO: The is no significant difference in the
average yield of the 3 different variety of crops
HI: The is significant difference in the
average yield of the 3 different variety of crops

Flowchart

,ANALYZE

COMPARE

ONE WAY ANOVA


(F TEST)

DEPENDENT
i ophioNs vdST
YIELD VARIETY- {fDYSCRlPilVh DUNC+N

CONTINUE

OK
OUT PUT

Yield N Mean Standard Standard Minimum


Deviation Error 95% confidence
interval for mean
Lower Upper
Bound Bound

Variety
i 5 20.00 2.550 1.140 16,83 23.17 16 23

Variety 2 5 19.00 3.808 1.703 14.27 23.73 15 25

Variety 3 5 27.00 3.742 1.673 22.35 31.65 22 32

Total 15 22.00 4.855 1.254 19.31 24.69 15 32

Inference:
Since p value (0.516)is greater than 0.05 at 5 percent level of significance we accept
the null
hypothesis and reject the alternate hypothesis. Ilence there is no significant
difference between
the aptitude marks obtained between batch 1 and batch 2 students.
REGRESSION ANALYSIS

Question:

A Company wants to assess the impact


of R&D expenditure on annual profits. The following
table gives the information of 8 years:

YEAR R&D Expenditure (in thousands) Annual Profit (in thousands)


2012 9
45
2013 7
2014 5
2015 10 60
2016 4
30
2017 5 34
2018 3 25
2019 2 20
Estimate the regression equation and predict the
annual profit for the year,2020 for an allocated
sum of Rs.15,000 as R&D expenditure.

To perform regression analysis and to find ovt eke profit


for the year 2020

Proceedure:

1. Get into SPSS for windows for an untitled SPSS worksheet


2. Click the variable view, then enter the variable name, type of
data, width of the data, decimal
value, label for the variable name, values with possible options, scales and
measurement
3. Click the data view, the variable entered in variable view appear as column
names. Then enter
the data for the columns in a sequence until required.
4. Click file menu and then select save and enter a file name and save the file.
click Analyze and select regression and then select linear
6. A dialog box will appear in that Shift (profit) to dependent box and
then click ok

Result : Thus the required analysis is successfullyinformed


ANALYZE

REGRESSION

LINEAR 0K

PROFIT DEPENDENT VARIABLE

INDEPENDENT VARIABLE
OUTPUT:

Model Summa

Model R Square Adjusted R Std. Error of


S uare the Estimate

.80 4.241
1 .951a

a. Predictors: (Constant), expenditure

Coefficientsa
Model 1 Standardized
Unstandardized Coefficients
Coefficients
Std. Error Beta t Sig.

13.128 3.526 3.723 .010


(Constant)

Expenditure .567
4.266 .951 7.519 .000
a. Dependent Variable: profit

Expenditure will be taken as X (Independent variable) and Profit will be taken as Y (dependent)

Since profit to be estimated the regression equation will be like Y on X

From the output regression equation is Y = 13.13+4.27 X

Formula ( y= a+b*x)

To find the profit (y) if expenditure (x) is 15 ( thousands)

Profit (Y) = 13.13 + 4.27(15)


13.13 + 64.05 = 77.18
Profit = Rs.77,180 (77.18 x1000)

Thus the annual profit for the year 2020 is Rs.77,180 for an allocated sum of Rs. 15,000 as R&D
expenditure.
EXERCISE
CORRELATION
Question :

Calculate the Correlation coefficient between the age of husband and the agc of wife from the
following data.

Ageof 35 34 40 43 56 20 38
Husband
Age of Wife 32 30 31 32 53 20 33

Interpret the value of r.

Aim :
To find the correlation coefficient between the age of husband and wifc .
Procedure :
[Link] into SPSS for windows for an untitled SPSS workshcct
2,Click the variable view, then cntcr the variable name, type of data, width of the data, decimal
value, label for the variablc name, values with possible options, gcalcs and mcasurcmcnt.
[Link] the data view, the variable entered in variable view appear as column names. Then enter
the data for the columns in a sequence until required.
4. Click filc menu and then gcJcctsave and cntcr a filc name and gavethc file.
5. Click Analyze and then select Correlate and then select Bivariate
6. A dialog box will appear now Shift the Agc of husband and Age ofwifc into variabJcs box and
Click OK
7. Thus the output appears in the separate output file

Result: The value of Correlation coefficient ( r) is 0.937 .JIcnce there is high corrcJation
between age of husband and age of wife in the given data .
Flowchart :

ANALYZE

CORRELATE

BIVARIATE

VARIAIII„F..S OK
AGE OF
HUSBAND

AGE OF
WIFE

OUTPUT :

Correlations

AGE OF HUSBAND AGE OF WIFE


AGE OF HUSBAND Pearson Correlation 1.000 .93?
Sig. (2-tailod) .002
7.000 7
AGE OF WIFE Pearson Correlation 1.000
Sig. (2-tailed) .002
7 7.000
orrea on ssgr cana o ovo
EXERCISE 9

RANK CORRELATION
Question:

In a recitation program conducted in school Lor5thstandard students ten students participated Toe
students in the recitation contest and were ranked by three judges in the following order

First Judge SecondJudge Third Judge


1.0 2.0 3.0
4.0 6.0 7.0
6.0 5.0 4.0
3.0 4.0 5.0
2.0 7.0 10.0
9.0 10.0 8.0
7.0 9.0 9.0
8.0 3.0 2.0
10.0 8.0 6.0
5.0 1.0 1.0

Use the method of rank correlation coeffcient to determine which pair of judges have the
nearest approach in ranking the competitors?

Aim :
To perform rank correlation and find the relationship between three variables.

Procedure:

[Link] into SPSS for windows for an untitled SPSS worksheet


2. Click the variable view, then enter the variable name, type of data, width of the data, decimal
value, label for the variable name, values with possn)le options, scales and
measurement.
[Link] the data view, the variable entered in variable view appear as
column names. Then enter
the data for the columns in a sequence until required.
4. Click file menu and then select save and enter a file name and
savethe file.
5. Click Analyze and then select Correlate and then
select Bivariate
6. A dialog box will appear now Shift all the three variables
into the variables box and Click OK
and the output appears in the separate output
file
Flowchart:

ANALYZE

CORRELATE

BIVARIATE

VARIABLES OK

FIRST JUDGE

SECOND JUDGE

THIRD JUDGE

Output

Correla!ions
FIRST SECOND
FIRST Pearson Correlation THIRD
[Link] .503
Sig. (2-tailed)
.042
.138 .907
10.000 10
ECOND Pearson Correlation 10
.503
1.000
Sig. (2-tailed)
.138
.001
10 10.000
Pearson Correlation 10
.042
Sig. (2-tailed) 1.000
.907 .001
Correlation is significant 10 10
at the 0.0 level 10.000
ailed .
Result :
Since the rank correlation
and highest 0.867 among coefficient between second
the three coefficients. judge and third judge is
Therefore the second and positive
the third judge are
EXERCISE [Link]
CHI-SQUARE TEST
Question

Firm size Computer used in Computer used in Computer used in


Admin Desi Manufacture
Small 30 24 16
Medium 10 12 18
Large 32 18 10

Test the hypothesis that there is no association between size of the firm and use of compute

Aim :
To examine the significant association between the type of firm and usage of computers in the

Procedure:
l. Get into SPSS windows for an untitled SPSS worksheet
2. Click the variable view, then enter the variabie name, type of data, width of the data, decimal
value, label for the variable name, values with possible options, scales and measurement.
[Link] the data view, the variable entered in variable view appear as column names. Then enter
the data for the columns in a sequence until required.
4. Click file menu click save and enter a file name .Thus the data is saved
5. Click Analyze and select Descriptive statistics and the select Crosstabs.
6. A dialog box will appear, now select the one of the variable(Firm size) in 'rows' and another
one variable(Computer Usage) in 'column'
7. Now click Statistics button and select 'Chi-Square' and click continue
8. Now click Cells button and select 'Expected Frequency' and click continue and click OK
9. Thus the chi-square result is being appeared in the output window

Resu,t :
Flow Chart:
computer usage
Null hypothcnfs : There is no significant diffcrcnce between firm size and
computer usage
Altcrnnte hypothesis : There is a significant diffcrcncc between firm size and

FLOWCHART

ANALYSE

DESCRIPTIVE
STAflSTICS

CROSS TABS

Rows
FIRM SIZE
COMPUTER USAGE

COLUMNS

STATISTICS CHI-SQUARE CONTINUE OK

CELLS EXPECTED
FREQUENCY CONTINUE
OUTPUT

firm size • Computer usa e Crosstabulation


Computer usage
admin design manufacture Total

size small 24 16 70

Expected Count 29.6 222 18.1 70.0

mediu-n Count 10 12 18 40

Expected Count 16.9 12.7 10.4 40.0

large 18 10 60

Expected Count 25.4 19.1 15.5 60.0

Count 72 170

Expected Count 72.0 54.0 44.0 170.0

Chi-SquareTests
Asymp. Sig. (2-
Value df sided)

>eatsorjChi-Square 12.659
0.013
Note : Since 0.013 < 0.05 we
accept alternate hypothesis
Ratio 12.288 .015
ea-by-Linear Association 1.091 296
"o Va5dCases 170
a. O ceCs (80%) have expected count less than 5. The minimum expected
cant is 10.35.

RESULT:

Since p value (0.013 ) is less than < 0.05 we reject the null hypothesis and accept the alternate
hypothesis. There is significant difference between the firm size and usage of Computers in the

You might also like