| Type: | Package |
| Title: | Analysis of Design of Experiments for Biological Research |
| Version: | 0.1.0 |
| Description: | Performs analysis of popular experimental designs used in the field of biological research. The designs covered are completely randomized design, randomized complete block design, factorial completely randomized design, factorial randomized complete block design, split plot design, strip plot design and latin square design. The analysis include analysis of variance, coefficient of determination, normality test of residuals, standard error of mean, standard error of difference and multiple comparison test of means. The package has functions for transformation of data and yield data conversion. Some datasets are also added in order to facilitate examples. |
| License: | GPL-3 |
| Encoding: | UTF-8 |
| LazyData: | true |
| RoxygenNote: | 7.1.0 |
| Imports: | agricolae (≥ 1.3.3), stats (≥ 4.0.2) |
| Depends: | R (≥ 2.10) |
| NeedsCompilation: | no |
| Packaged: | 2020-07-08 11:19:58 UTC; Asus |
| Author: | Raj Popat [aut, cre], Kanthesh Banakara [aut] |
| Maintainer: | Raj Popat <popatrajc@gmail.com> |
| Repository: | CRAN |
| Date/Publication: | 2020-07-08 12:20:03 UTC |
Re-transform the Arc sine transformed data
Description
Re-transform the arc sine transformed data. When arc sine transformation is done, the mean of the treatments needs to be re-transformed for comparison.
Usage
arcsineretransform(mean.vector, type)
Arguments
mean.vector |
vector of mean which needs to be re-transformed |
type |
0 if data was in proportion prior to re-transformation, 1 if data was in percentage prior to re-transformation |
Value
Arc sine re-transformed vector
Examples
data<-c(60,63.43495,71.56505,78.46304)
#If data was in percentage prior to re-transformation
arcsineretransform(data,1)
#If data was in proportion prior to re-transformation
arcsineretransform(data,0)
Arc sine transformation of the numeric vector
Description
The function divide values by 100, does square root and than sin inverse of each values of vector. If any of the values of a vector is 0 or 100, it is replaced by 1/4n or 100-(1/4n), respectively.
Usage
arcsinetransform(numeric.vector, type, n)
Arguments
numeric.vector |
data vector to be transformed |
type |
0 if data is in percentage and 1 if data is in proportion |
n |
is the number of units upon which the percentage/proportion data is based |
Value
Arc sine transformed data
Examples
vector<-c(23,0,29.6,35.6,33,35.6,10.5,100)
# Arc sine trnasformation for percentage data and n=10
arcsinetransform(vector,0,10)
Convert the data frame into list of numeric nature
Description
Convert the data frame into list of numeric nature
Usage
convert(data1)
Arguments
data1 |
data-frame to be converted into list |
Value
list of numeric vectors
Analysis of Completely Randomized Design
Description
The function gives ANOVA, R-square of the model, normality testing of residuals, SEm (standard error of mean), SEd (standard error of difference), interpretation of ANOVA results and multiple comparison test for means
Usage
crd(data, trt.vector, MultipleComparisonTest)
Arguments
data |
dependent variables |
trt.vector |
vector containing treatments |
MultipleComparisonTest |
0 for no test, 1 for LSD test, 2 for Duncan test and 3 for HSD test |
Value
ANOVA, interpretation of ANOVA, R-square, normality test result, SEm, SEd and multiple comparison test result
Examples
data<-data.frame(Treatments=c("T1","T2","T3","T4","T5","T6","T7","T1","T2","T3","T4","T5","T6",
"T7","T1","T2","T3","T4","T5","T6","T7"),
yield=c(25,21,21,18,25,28,24,25,24,24,16,21,20,17,16,19,14,15,13,11,25),
height=c(130,120,125,135,139,140,145,136,129,135,150,152,140,148,130,135,145,160,145,130,160))
#CRD analysis with LSD test for yield only
crd(data[2],data$Treatments,1)
#CRD analysis with LSD test for both yield and height
crd(data[2:3],data$Treatments,1)
Data of Factorial Experiment
Description
The data consists of three factors nitrogen, phosphorus and Potassium, replication and two dependent variables yield and plant height. The data is generated manually.
Usage
factorialdata
Format
The data has 6 columns and 36 rows
- Nitrogen
Consist sequence of two nitrogen levels n0 and n1
- Phosphorus
Consist sequence of two phosphorus levels p0 and p1
- Potassium
Consist sequence of two potassium levels k0 and k1
- Replication
Contains replication which has three levels
- Yield
Yield as dependent variable
- Plant Height
Plant height as dependent variable
Analysis of Factorial Completely Randomized Design for 2 factors
Description
The function gives ANOVA, R-square of the model, Normality testing of residuals, SEm (standard error of mean), SEd (standard error of difference), interpretation of ANOVA results and multiple comparison test for means
Usage
fcrd2fact(data, fact.A, fact.B, Multiple.comparison.test)
Arguments
data |
dependent variables |
fact.A |
vector containing levels of first factor |
fact.B |
vector containing levels of second factor |
Multiple.comparison.test |
0 for no test, 1 for LSD test, 2 for Duncan test and 3 for HSD test |
Value
ANOVA, interpretation of ANOVA, R-square, normality test result, SEm, SEd and multiple comparison test result for both the factors as well as interaction.
Examples
data(factorialdata)
#Analysis of Factorial Completely Randomized design along with Dunccan test for Yield only
fcrd2fact(factorialdata[5],factorialdata$Nitrogen,factorialdata$Phosphorus,2)
#Analysis of Factorial Completely Randomized design along with Dunccan test for Yield & Plant Height
fcrd2fact(factorialdata[5:6],factorialdata$Nitrogen,factorialdata$Phosphorus,2)
Analysis of Factorial Completely Randomized Design for 3 factors
Description
The function gives ANOVA, R-square of the model, normality testing of residuals, SEm (standard error of mean), SEd (standard error of difference), interpretation of ANOVA results and multiple comparison test for means.
Usage
fcrd3fact(data, fact.A, fact.B, fact.C, Multiple.comparison.test)
Arguments
data |
dependent variables |
fact.A |
vector containing levels of first factor |
fact.B |
vector containing levels of second factor |
fact.C |
vector containing levels of third factor |
Multiple.comparison.test |
0 for no test, 1 for LSD test, 2 for Duncan test and 3 for HSD test |
Value
ANOVA, interpretation of ANOVA, R-square, normality test result, SEm, SEd and multiple comparison test result for both the factors as well as interaction.
Examples
data(factorialdata)
#FCRD analysis along with dunccan test for two dependent var.
fcrd3fact(factorialdata[5:6],factorialdata$Nitrogen,
factorialdata$Phosphorus,factorialdata$Potassium,2)
Analysis of Factorial Randomized Block Design for 2 factors
Description
The function gives ANOVA, R-square of the model, normality testing of residuals, SEm (standard error of mean), SEd (standard error of difference), interpretation of ANOVA results and multiple comparison test for means.
Usage
frbd2fact(data, replicationvector, fact.A, fact.B, Multiple.comparison.test)
Arguments
data |
dependent variables |
replicationvector |
vector containing replications |
fact.A |
vector containing levels of first factor |
fact.B |
vector containing levels of second factor |
Multiple.comparison.test |
0 for no test, 1 for LSD test, 2 for Duncan test and 3 for HSD test |
Value
ANOVA, interpretation of ANOVA, R-square, normality test result, SEm, SEd and multiple comparison test results for both the factors as well as interaction.
Examples
data(factorialdata)
#FRBD analysis along with dunccan test for two dependent var.
frbd2fact(factorialdata[5:6],factorialdata$Replication,
factorialdata$Nitrogen,factorialdata$Phosphorus,2)
Analysis of Factorial Randomized Block Design for 3 factors
Description
The function gives ANOVA, R-square of the model, normality testing of residuals, SEm (standard error of mean), SEd (standard error of difference), interpretation of ANOVA results and multiple comparison test for means.
Usage
frbd3fact(
data,
replicationvector,
fact.A,
fact.B,
fact.C,
Multiple.comparison.test
)
Arguments
data |
dependent variables |
replicationvector |
vector containing replications |
fact.A |
vector containing levels of first factor |
fact.B |
vector containing levels of second factor |
fact.C |
vector containing levels of third factor |
Multiple.comparison.test |
0 for no test, 1 for LSD test, 2 for Duncan test and 3 for HSD test |
Value
ANOVA, interpretation of ANOVA, R-square, normality test result, SEm, SEd and multiple comparison test result for the factors as well as the interaction.
Examples
data(factorialdata)
#FRBD analysis along with dunccan test for two dependent var.
frbd3fact(factorialdata[5:6],factorialdata$Replication,factorialdata$Nitrogen,
factorialdata$Phosphorus,factorialdata$Potassium,2)
Re-transform the log transformed data
Description
Re-transform the log transformed data. When log transformation is done, the mean of the treatments needs to be re-transformed for comparison.
Usage
logretransform(transformed.mean, if.zero.present)
Arguments
transformed.mean |
vector of mean which needs to be re-transformed |
if.zero.present |
0 if zero was present in the data prior to transformation of data. 1 if zero was absent in the data prior to transformation |
Value
Log re-transformed values
Examples
vector<-c(0,2.004,1.114,1.491,1.431,1.415,1.845)
#Re-transformation of data with zero present in data prior to transformation
logretransform(vector,0)
Log transformation of the numeric vector
Description
The function carries out log with base 10 transformation of each values of vector. If one of values of a vector is 0, 1 is added to each observation. Log transformation is carried out for the data when variance is proportional to square of the mean and treatment effects are multiplicative in nature.
Usage
logtransform(numeric.vector)
Arguments
numeric.vector |
data vector to be transformed |
Value
A list of
-
Ratio- A ratio of maximum and minimum values of the data -
LogTransformedVector- A vector of the transformed data -
Comment- A comment about zero being present in data or not
Examples
vector<-c(100,0,120,1000,52,30,60)
logtransform(vector)
Analysis of Latin Square Design
Description
The function gives ANOVA, R-square of the model, normality testing of residuals, SEm (standard error of mean), SEd (standard error of difference), interpretation of ANOVA results and multiple comparison test for means.
Usage
lsd(data, treatmentvector, row, column, MultipleComparisonTest)
Arguments
data |
dependent variables |
treatmentvector |
vector containing treatments |
row |
vector for rows |
column |
vector for columns |
MultipleComparisonTest |
0 for no test, 1 for LSD test, 2 for Duncan test and 3 for HSD test |
Value
ANOVA, interpretation of ANOVA, R-square, normality test result, SEm, SEd and multiple comparison test result
Examples
data(lsddata)
#LSD analysis with LSD test for Yield only
lsd(lsddata[4],lsddata$Treatment,lsddata$Row,lsddata$Column,1)
#LSD analysis with LSD test for Yield and Plant Height
lsd(lsddata[4:5],lsddata$Treatment,lsddata$Row,lsddata$Column,1)
Data for Latin Square Design
Description
The data consists of Rows, Columns, Treatments and two dependent variables Yield and Plant Height. The data is generated manually.
Usage
lsddata
Format
The data has 5 columns and 25 rows
- Row
Consist sequence of rows. Row consists of 5 levels
- Column
Consist sequence of column. Column consists of 5 levels
- Treatment
Consist sequence of treatments. There are 5 treatments A, B, C, D & E
- Yield
Yield as dependent variable
- Plant Height
Plant height as dependent variable
Analysis of Randomized Complete Block Design
Description
The function gives ANOVA, R-square of the model, normality testing of residuals, SEm (standard error of mean), SEd (standard error of difference), interpretation of ANOVA results and multiple comparison test for means.
Usage
rcbd(data, treatmentvector, replicationvector, MultipleComparisonTest)
Arguments
data |
dependent variables |
treatmentvector |
vector containing treatments |
replicationvector |
vector containing replications |
MultipleComparisonTest |
0 for no test, 1 for LSD test, 2 for Duncan test and 3 for HSD test |
Value
ANOVA, interpretation of ANOVA, R-square, normality test result, SEm, SEd and multiple comparison test result
Examples
data<-data.frame(GFY=c(16,13,14,16,16,17,16,17,16,16,17,16,15,15,15,13,15,14,
16,14,15,14,15,17,18,15,15,15,14,14,14,14,15,15,13,15,14,14,13,13,13,12,15,12,15),
DMY=c(5,5,6,5,6,7,6,8,6,9,8,7,5,5,5,4,6,5,8,5,5,5,4,6,6,5,5,6,6,6,5,5,5,5,5,6,5,5,5,4,5,4,5,5,5),
Rep=rep(c("R1","R2","R3"),each=15),
Trt=rep(c("T1","T2","T3","T4","T5","T6","T7","T8","T9","T10","T11","T12","T13","T14","T15"),3))
#' #RCBD analysis with duncan test for GFY only
rcbd(data[1],data$Trt,data$Rep,2)
#RCBD analysis with duncan test for both GFY and DMY
rcbd(data[1:2],data$Trt,data$Rep,2)
Data for Split plot Design
Description
The data consists of replication, date of sowing (as main-plot), varieties (as sub-plot) and two dependent variables yield and plant height. The data is generated manually.
Usage
splitdata
Format
The data has 5 columns and 36 rows
- Replication
Consist sequence of replications. Replications consists of 3 levels
- Date of Sowing
Consist sequence of levels of date of sowing as Main-plot. Date of sowing consists of 2 levels
- Varities
Consist sequence of levels of varities as Sub-plot. Varities consist of 6 levels
- Yield
Yield as dependent variable
- Plant Height
Plant height as dependent variable
Analysis of Split plot design
Description
The function gives ANOVA, R-square of the model, normality testing of residuals, SEm (standard error of mean), SEd (standard error of difference), interpretation of ANOVA results and multiple comparison test for means.
Usage
splitplot(data, block, main.plot, sub.plot, mean.comparison.test)
Arguments
data |
dependent variables |
block |
vector containing replications |
main.plot |
vector containing main-plot levels |
sub.plot |
vector containing sub-plot levels |
mean.comparison.test |
0 for no test, 1 for LSD test, 2 for Dunccan test and 3 for HSD test |
Value
ANOVA, interpretation of ANOVA, R-square, normality test result, SEm, SEd and multiple comparison test result
Examples
data(splitdata)
#Using Date of sowing as Main-plot factor and varieties as sub-plot factor and using LSD test
#Split plot analysis with LSD test for Yield
splitplot(splitdata[4],splitdata$Replication,splitdata$Date_of_Sowing,splitdata$Varities,1)
#Split plot analysis with LSD test for both Yield and Plant Height
splitplot(splitdata[4:5],splitdata$Replication,splitdata$Date_of_Sowing,splitdata$Varities,1)
Re-transform the square root transformed data
Description
Retransform the square root transformed data. When square root transformation is done, the mean of the treatments needs to be re-transformed for comparison.
Usage
sqrtretransform(transformed.mean, if.zero.present)
Arguments
transformed.mean |
vector of mean which needs to be re-transformed |
if.zero.present |
0 if zero was present in the data prior to transformation of data. 1 if zero was absent in the data prior to transformation |
Value
Square root re-transformed vector
Examples
vector<-c(19,10,30,60,50,10,5)
#Square root re-transform and zero was absent in the data prior to transformation
sqrtretransform(vector,1)
Square root transformation of the numeric vector
Description
The function carries out square root transformation of each values of vector. If one of values of a vector is 0, 0.5 is added to each observation.
Usage
sqrttransform(numeric.vector)
Arguments
numeric.vector |
data vector to be transformed |
Value
Square root transformed data
Examples
vector<-c(0,25,36,6,9,25,70)
sqrttransform(vector)
Analysis of Strip plot design
Description
The function gives ANOVA, R-square of the model, normality testing of residuals, SEm (standard error of mean), SEd (standard error of difference), interpretation of ANOVA results and multiple comparison test for means
Usage
stripplot(data, block, column, row, mean.comparison.test)
Arguments
data |
dependent variables |
block |
vector containing replications |
column |
vector containing column strip levels |
row |
vector containing row strip levels |
mean.comparison.test |
0 for no test, 1 for LSD test, 2 for Duncan test and 3 for HSD test |
Value
ANOVA, interpretation of ANOVA, R-square, normality test result, SEm, SEd and multiple comparison test result
Examples
data(splitdata)
#Split data is used for sake of demonstration
#Using Date of sowing as Column factor and varieties as Row factor and using LSD test for Yield only
stripplot(splitdata[4],splitdata$Replication,splitdata$Date_of_Sowing,splitdata$Varities,1)
#Using Date of sowing as Column factor and varieties as Row factor and using LSD test for both var.
stripplot(splitdata[4:5],splitdata$Replication,splitdata$Date_of_Sowing,splitdata$Varities,1)
Convert the yield data of plot into different units
Description
The function converts the yield data of plot into qtl/ha, tonnes/ha, qtl/acre or tonnes/acre depending on the option chosen.
Usage
yieldconvert(yield.in.kg, length.of.plot, width.of.plot, choose.convert.to)
Arguments
yield.in.kg |
yield data in kilograms |
length.of.plot |
length of plot in m |
width.of.plot |
width of the plot in m |
choose.convert.to |
0 for qtl/ha, 1 for tonnes/ha, 2 for qtl/acre and 3 for tonnes/acre |
Value
converted yield
Examples
#Convert yield vector obtained from 10m x 5m plot into different forms
yield<-c(10,15,12,16,19,25,30,25,11)
#For converting into qtl/ha
yieldconvert(yield,10,5,0)
#For converting into tonnes/ha
yieldconvert(yield,10,5,1)
#For converting into qtl/acre
yieldconvert(yield,10,5,2)
#For converting into tonnes/acre
yieldconvert(yield,10,5,3)