Biology 201

An example showing the difference in usage between unpaired and paired t-tests

Hibernating Bears

Do they loose weight?

"Do they loose weight?" This is an ambiguous question, in and of itself. It could mean, "Do individual bears loose weight?" or "Is the mass of these seven bears significantly different after dormancy when compared with their masses prior to dormancy?" Make sure you know which question you are addressing!

Bear Number

mass before (kg)

mass after (kg)

1

298

276

2

152

140

3

166

151

4

305

279

5

90

81

6

121

107

7

225

199

mean

193.9

176.1

 Numbers are easily crunched in a spreadsheet set up like this: 

Bear Number

mass before dormancy (kg)

mass2

 

mass after dormancy (kg)

mass2

 

mass difference (kg)

(X-Y) 2

1

298

88804

 

276

76176

 

22

484

2

152

23104

 

140

19600

 

12

144

3

166

27556

 

151

22801

 

15

225

4

305

93025

 

279

77841

 

26

676

5

90

8100

 

81

6561

 

9

81

6

121

14641

 

107

11449

 

14

196

7

225

50625

 

199

39601

 

26

676

mean

193.9

 

 

176.1

 

 

17.7

 

sum

1357

305855

 

1233

254029

 

124

2482

sum2

1841449

 

 

1520289

 

 

15376

 

SS

42791

 

 

36845

 

 

285

 

s2

7132

 

 

6141

 

 

48

 

 

 

 

 

 

s

(SD)

6.9

 

t Student's

0.4068

 

 

 

 

 

t paired

6.795

 

Has the average mass of the bears changed? i.e. Is there a significant difference between the average masses of the bears before and after the dormancy? Use the t-statistic from the Student's t-test to determine this.

Is there a tendency for individual bears to lose weight during dormancy? Use the paired t-test to determine this.

What are the degrees of freedom for this example? Look up the critical value of t, using an α=0.05.

 

 

  


 


Home
R.F. Lauff
Department of Biology
St. Francis Xavier University
Antigonish, NS
Canada B2G 2W5