In: Statistics and Probability
3. The following data are weights of food (in kilograms) consumed per day by adult deer collected at different times of the year.
Month Weight of food consumed (kg)
Feb. 4.7, 4.9, 5.0, 4.8, 4.7
Mar. 4.6, 4.4, 4.3, 4.4, 4.1, 4.2
Oct. 4.8, 4.7, 4.6, 4.4, 4.7, 4.8
Dec. 4.9, 5.2, 5.4, 5.1, 5.6
(a) Is there any difference in the food consumption for all the months studied?
(b) If food consumption is not the same for all the months, which ones are different (you must show the test used and results for interpretation)?
Solve using Excel, go to Data, select Data Analysis, choose Anova: single factor and groupp by rows
a) H0: µ1 = µ2 = µ3 = µ4, There is no difference in the food consumption for all the months studied
H1: There is a difference in the food consumption for all the months studied
Anova: Single Factor | ||||||
SUMMARY | ||||||
Groups | Count | Sum | Average | Variance | ||
Feb | 5 | 24.100 | 4.820 | 0.017 | ||
Mar | 6 | 26.000 | 4.333 | 0.031 | ||
Oct | 6 | 28.000 | 4.667 | 0.023 | ||
Dec | 5 | 26.200 | 5.240 | 0.073 | ||
ANOVA | ||||||
Source of Variation | SS | df | MS | F | P-value | F crit |
Between Groups | 2.307 | 3 | 0.769 | 22.084 | 0.000 | 3.160 |
Within Groups | 0.627 | 18 | 0.035 | |||
Total | 2.933 | 21 |
Since p-value is less than 0.05, we reject the null hypotesis and conclude that there is a difference in the food consumption for all the months studied.
b) To detect difference in pairs, conduct t-tests on Excel
using Excel, go to Data, select Data Analysis, choose t-Test: Two-Sample Assuming Unequal Variances
For (Feb, Mar)
H0: µ1 = µ2
H1: µ1 ≠ µ2
t-Test: Two-Sample Assuming Unequal Variances | ||
Feb | Mar | |
Mean | 4.82 | 4.333333 |
Variance | 0.017 | 0.030667 |
Observations | 5 | 6 |
Hypothesized Mean Difference | 0 | |
df | 9 | |
t Stat | 5.275194 | |
P(T<=t) one-tail | 0.000255 | |
t Critical one-tail | 1.833113 | |
P(T<=t) two-tail | 0.00051 | |
t Critical two-tail | 2.262157 |
Since p-value = 0.00051 is more than 0.05, we reject the null hypothessis and conclude that µ1 ≠ µ2
For (Feb, Oct),
H0: µ1 = µ3
H1: µ1 ≠ µ3
t-Test: Two-Sample Assuming Unequal Variances | ||
Feb | Oct | |
Mean | 4.82 | 4.666667 |
Variance | 0.017 | 0.022667 |
Observations | 5 | 6 |
Hypothesized Mean Difference | 0 | |
df | 9 | |
t Stat | 1.809846 | |
P(T<=t) one-tail | 0.05188 | |
t Critical one-tail | 1.833113 | |
P(T<=t) two-tail | 0.10376 | |
t Critical two-tail | 2.262157 |
Since p-value = 0.10376 is more than 0.05, we do not reject the null hypothessis and conclude that µ1 = µ3
For (Feb, Mar)
H0: µ1 = µ4
H1: µ1 ≠ µ4
t-Test: Two-Sample Assuming Unequal Variances | ||
Feb | Dec | |
Mean | 4.82 | 5.24 |
Variance | 0.017 | 0.073 |
Observations | 5 | 5 |
Hypothesized Mean Difference | 0 | |
df | 6 | |
t Stat | -3.1305 | |
P(T<=t) one-tail | 0.010156 | |
t Critical one-tail | 1.94318 | |
P(T<=t) two-tail | 0.020312 | |
t Critical two-tail | 2.446912 |
Since p-value = 0.020312 is less than 0.05, we reject the null hypothessis and conclude that µ1 ≠ µ4
For (Mar, Oct),
H0: µ2 = µ3
H1: µ2 ≠ µ3
t-Test: Two-Sample Assuming Unequal Variances | ||
Mar | Oct | |
Mean | 4.333333 | 4.666667 |
Variance | 0.030667 | 0.022667 |
Observations | 6 | 6 |
Hypothesized Mean Difference | 0 | |
df | 10 | |
t Stat | -3.53553 | |
P(T<=t) one-tail | 0.002698 | |
t Critical one-tail | 1.812461 | |
P(T<=t) two-tail | 0.005397 | |
t Critical two-tail | 2.228139 |
Since p-value = 0.005397 is more than 0.05, we do not reject the null hypothessis and conclude that µ2 = µ3
For (Mar, Dec)
H0: µ2 = µ4
H1: µ2 ≠ µ4
t-Test: Two-Sample Assuming Unequal Variances | ||
Mar | Dec | |
Mean | 4.333333 | 5.24 |
Variance | 0.030667 | 0.073 |
Observations | 6 | 5 |
Hypothesized Mean Difference | 0 | |
df | 7 | |
t Stat | -6.45791 | |
P(T<=t) one-tail | 0.000174 | |
t Critical one-tail | 1.894579 | |
P(T<=t) two-tail | 0.000348 | |
t Critical two-tail | 2.364624 |
Since p-value = 0.00003 is less than 0.05, we reject the null hypothessis and conclude that µ2 ≠ µ4
For (Mar, Oct),
H0: µ2 = µ3
H1: µ2 ≠ µ3
t-Test: Two-Sample Assuming Unequal Variances | ||
Mar | Oct | |
Mean | 4.333333 | 4.666667 |
Variance | 0.030667 | 0.022667 |
Observations | 6 | 6 |
Hypothesized Mean Difference | 0 | |
df | 10 | |
t Stat | -3.53553 | |
P(T<=t) one-tail | 0.002698 | |
t Critical one-tail | 1.812461 | |
P(T<=t) two-tail | 0.005397 | |
t Critical two-tail | 2.228139 |
Since p-value = 0.005397 is more than 0.05, we do not reject the null hypothessis and conclude that µ2 = µ3
For (Oct, Dec)
H0: µ3 = µ4
H1: µ3 ≠ µ4
t-Test: Two-Sample Assuming Unequal Variances | ||
Oct | Dec | |
Mean | 4.666667 | 5.24 |
Variance | 0.022667 | 0.073 |
Observations | 6 | 5 |
Hypothesized Mean Difference | 0 | |
df | 6 | |
t Stat | -4.22922 | |
P(T<=t) one-tail | 0.002752 | |
t Critical one-tail | 1.94318 | |
P(T<=t) two-tail | 0.005505 | |
t Critical two-tail | 2.446912 |
Since p-value = 0.0005 is less than 0.05, we reject the null hypothessis and conclude that µ3 ≠ µ4