dic.ad: Analysis: Completely randomized design with an additional...

View source: R/dic_ad.R

dic.adR Documentation

Analysis: Completely randomized design with an additional treatment for quantitative factor

Description

Statistical analysis of experiments conducted in a completely randomized with an additional treatment and balanced design with a factor considering the fixed model.

Usage

dic.ad(
  trat,
  response,
  responsead,
  grau = 1,
  norm = "sw",
  homog = "bt",
  alpha.f = 0.05,
  theme = theme_classic(),
  ylab = "response",
  xlab = "independent",
  family = "sans",
  posi = "top",
  pointsize = 4.5,
  linesize = 0.8,
  width.bar = NA,
  point = "mean_sd"
)

Arguments

trat

Numerical or complex vector with treatments

response

Numerical vector containing the response of the experiment.

responsead

Numerical vector with additional treatment responses

grau

Degree of polynomial in case of quantitative factor (default is 1)

norm

Error normality test (default is Shapiro-Wilk)

homog

Homogeneity test of variances (default is Bartlett)

alpha.f

Level of significance of the F test (default is 0.05)

theme

ggplot2 theme (default is theme_classic())

ylab

Variable response name (Accepts the expression() function)

xlab

Treatments name (Accepts the expression() function)

family

Font family

posi

Legend position

pointsize

Point size

linesize

line size (Trendline and Error Bar)

width.bar

width of the error bars of a regression graph.

point

Defines whether to plot mean ("mean"), mean with standard deviation ("mean_sd" - default) or mean with standard error (default - "mean_se"). For quali=FALSE or quali=TRUE.

Value

The table of analysis of variance, the test of normality of errors (Shapiro-Wilk ("sw"), Lilliefors ("li"), Anderson-Darling ("ad"), Cramer-von Mises ("cvm"), Pearson ("pearson") and Shapiro-Francia ("sf")), the test of homogeneity of variances (Bartlett ("bt") or Levene ("levene")), the test of independence of Durbin-Watson errors, adjustment of regression models up to grade 3 polynomial. The function also returns a standardized residual plot.

Note

In some experiments, the researcher may study a quantitative factor, such as fertilizer doses, and present a control, such as a reference fertilizer, treated as a qualitative control. In these cases, there is a difference between considering only the residue in the unfolding of the polynomial, removing or not the qualitative treatment, or since a treatment is excluded from the analysis. In this approach, the residue used is also considering the qualitative treatment, a method similar to the factorial scheme with additional control.

Author(s)

Gabriel Danilo Shimizu, shimizu@uel.br

Leandro Simoes Azeredo Goncalves

Rodrigo Yudi Palhaci Marubayashi

Examples

datadicad=data.frame(doses = c(rep(c(1:5),e=3)),
                     resp = c(3,4,3,5,5,6,7,7,8,4,4,5,2,2,3))
with(datadicad,dic.ad(doses, resp, rnorm(3,6,0.1),grau=2))

AgroR documentation built on May 29, 2024, 4:18 a.m.