\[H_0: \mu_1= \mu_2 = \mu_3 = \mu_4\] \[ H_{alt}: \mbox{at least one mean is different}\] where the means \((\mu_{i..4})\) represent the means of the four different groups.
| fertilizer | height | |
|---|---|---|
| F0:6 | Min. :19.50 | |
| F1:6 | 1st Qu.:22.50 | |
| F2:6 | Median :27.25 | |
| F3:6 | Mean :26.17 | |
| NA | 3rd Qu.:28.75 | |
| NA | Max. :32.00 |
The summary table for the separate groups (F0, F1, F2, and F3) shows the arithmetic mean for plant height, medium plant height, variance, and standard deviation. The plot of group means is also shown.
| fertilizer | mean_height | median_height | var_height | stdev_height |
|---|---|---|---|---|
| F0 | 21.00000 | 21.00 | 1.000000 | 1.000000 |
| F1 | 28.60000 | 28.30 | 5.940000 | 2.437212 |
| F2 | 25.86667 | 26.25 | 3.606667 | 1.899123 |
| F3 | 29.20000 | 29.35 | 1.660000 | 1.288410 |
A one-way analysis of variance showed that there were significant differences in plant height means between the groups at the \(p<.05\) level, \(F(3,20)=27.46, p<.001\). Follow-up comparisons were performed.
## Df Sum Sq Mean Sq F value Pr(>F)
## fertilizer 3 251.44 83.81 27.46 2.71e-07 ***
## Residuals 20 61.03 3.05
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## (Intercept) fertilizerF1 fertilizerF2 fertilizerF3
## 21.000000 7.600000 4.866667 8.200000
Post hoc comparisons were performed using the Tukey HSD test. The results showed that all three fertilizer groups (F1, F2, and F3) were significantly higher than the group with no fertilizer (F0), at the \(p<.05\) level. The results also showed that formula F3 was significantly higher than formula F2 at the \(p<.05\) level. The mean plant height differences between formulas F1 and F2 were not significant at the p<.05 level. The mean height differences between formulas F1 and F3 were also not significant at the p<.05 level.
##
## Simultaneous Tests for General Linear Hypotheses
##
## Multiple Comparisons of Means: Tukey Contrasts
##
##
## Fit: aov(formula = height ~ fertilizer)
##
## Linear Hypotheses:
## Estimate Std. Error t value Pr(>|t|)
## F1 - F0 == 0 7.600 1.009 7.535 <0.001 ***
## F2 - F0 == 0 4.867 1.009 4.825 <0.001 ***
## F3 - F0 == 0 8.200 1.009 8.130 <0.001 ***
## F2 - F1 == 0 -2.733 1.009 -2.710 0.060 .
## F3 - F1 == 0 0.600 1.009 0.595 0.932
## F3 - F2 == 0 3.333 1.009 3.305 0.017 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## (Adjusted p values reported -- single-step method)
## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = height ~ fertilizer)
##
## $fertilizer
## diff lwr upr p adj
## F1-F0 7.600000 4.7770648 10.42293521 0.0000016
## F2-F0 4.866667 2.0437315 7.68960188 0.0005509
## F3-F0 8.200000 5.3770648 11.02293521 0.0000005
## F2-F1 -2.733333 -5.5562685 0.08960188 0.0598655
## F3-F1 0.600000 -2.2229352 3.42293521 0.9324380
## F3-F2 3.333333 0.5103981 6.15626854 0.0171033
Data Source: From Penn State University stats project data.
03/2021