Analysis of variance for a completely randomized design with standard errors, critical difference, coefficient of variation, post-hoc letter groupings and assumption checks. Unequal replication is handled exactly.
Usage
anova_crd(
data,
response,
treatment,
posthoc = "lsd",
alpha = 0.05,
verbose = TRUE,
control = NULL
)Arguments
- data
Data frame containing the experimental data.
- response
Name of the response variable (character).
- treatment
Name of the treatment column.
- posthoc
Post-hoc test(s):
"lsd","duncan","tukey","snk","scheffe","bonferroni","dunnett", a vector of these,"all"or"none". The first method supplies the letters in the means table.- alpha
Significance level for the CD and the post-hoc tests (default 0.05).
- verbose
Logical. If TRUE (default), prints the report.
- control
Control level for Dunnett's test (default: first level).
Value
An object of class aridagri_anova with
anova_table, model, grand_mean, cv,
ms_error, df_error, se_mean (SEm),
se_diff (SEd), cd (at alpha), cd_5,
cd_1, treatment_means (mean, SD, N, SE and letters),
posthoc, diagnostics, significance and
balanced.
Details
SEm = sqrt(MSE / r), SEd = sqrt(2 MSE / r) and CD = t(alpha/2, error df) x SEd, where r is the number of replications (the harmonic mean when replication is unequal; the letters then use the exact SE of each pair). CV = 100 sqrt(MSE) / grand mean. Rows with a missing response are dropped with a message.
References
Gomez, K.A. and Gomez, A.A. (1984). Statistical Procedures for Agricultural Research, 2nd ed. John Wiley and Sons, New York.
Examples
data <- data.frame(
treatment = rep(c("T1", "T2", "T3", "T4"), each = 5),
yield = c(1210, 1185, 1242, 1198, 1225, 1340, 1362, 1318, 1375, 1350,
1105, 1090, 1122, 1131, 1098, 1445, 1462, 1430, 1471, 1452)
)
res <- anova_crd(data, response = "yield", treatment = "treatment",
posthoc = "duncan")
#>
#> ======================================================================
#> COMPLETELY RANDOMIZED DESIGN (CRD)
#> ======================================================================
#> Response : yield
#> Treatment : treatment
#> Treatments : 4
#> Observations : 20
#>
#> Analysis of variance
#> Source Df Sum Sq Mean Sq F value Pr(>F)
#> Treatment 3 340702.15 113567.38 301.300 0.0000 ***
#> Error 16 6030.80 376.92
#> Total 19 346732.95
#> Signif. codes: *** p < 0.001, ** p < 0.01, * p < 0.05, NS not significant
#>
#> Grand mean = 1280.550 CV (%) = 1.516 SEm = 8.682 SEd = 12.279 CD (5%) = 26.030
#>
#> Treatment means (Duncan's MRT, alpha = 0.05)
#> Treatment Mean SD N SE Group
#> T4 1452.00 15.764 5 8.682 a
#> T2 1349.00 21.726 5 8.682 b
#> T1 1212.00 22.349 5 8.682 c
#> T3 1109.20 16.962 5 8.682 d
#> Means sharing a letter do not differ significantly.
res$treatment_means
#> Treatment Mean SD N SE Group
#> 1 T4 1452.0 15.76388 5 8.682454 a
#> 2 T2 1349.0 21.72556 5 8.682454 b
#> 3 T1 1212.0 22.34950 5 8.682454 c
#> 4 T3 1109.2 16.96172 5 8.682454 d