Type:	Package
Title:	Power Analysis and Sample Size Calculation for Semi-Field Vector Control Trials
Version:	1.0.3
Maintainer:	Andrea M. Kipingu <akipingu@ihi.or.tz>
Description:	Uses simulations from generalized linear mixed-effects models to incorporate random effects across multiple sources and levels of variation, and a dispersion parameter to account for overdispersion and capture unexplained variability. Covers design scenarios for both short-term and long-term trials evaluating the impact of single or combined vector control interventions. Methods build on Kipingu et al. (2025) <doi:10.1186/s12936-025-05454-y> and Johnson et al. (2015) <doi:10.1111/2041-210X.12306>.
License:	MIT + file LICENSE
Encoding:	UTF-8
RoxygenNote:	7.3.3
Imports:	lme4, parallel, dplyr, ggplot2, stats
Suggests:	rmarkdown, testthat (≥ 3.0.0), waldo, devtools
Config/testthat/edition:	3
URL:	https://github.com/akipingu/simpowa
BugReports:	https://github.com/akipingu/simpowa/issues
NeedsCompilation:	no
Packaged:	2026-03-24 19:31:07 UTC; dr.kipingu
Author:	Andrea M. Kipingu [aut, cre] (affiliation: Environmental Health & Ecological Sciences Department, Ifakara Health Institute, P.O. Box 78 373, Dar es Salaam, Tanzania), Mafalda Viana [aut] (affiliation: School of Biodiversity, One Health and Veterinary Medicine, University of Glasgow, Graham Kerr Building, Glasgow, G12 8QQ, UK), Paul C.D. Johnson [aut] (affiliation: School of Biodiversity, One Health & Veterinary Medicine, University of Glasgow, Wolfson Link Building, Glasgow, G12 8QQ, UK), Samson S. Kiware [aut] (affiliation: Environmental Health & Ecological Sciences Department, Ifakara Health Institute, P.O. Box 78 373, Dar es Salaam, Tanzania)
Repository:	CRAN
Date/Publication:	2026-03-29 16:00:02 UTC

Power Analysis and Sample Size Calculation for Semi-Field Vector Control Trials

Description

Uses simulations from generalized linear mixed-effects models to incorporate random effects across multiple sources and levels of variation, and a dispersion parameter to account for overdispersion and capture unexplained variability. Covers design scenarios for both short-term and long-term trials evaluating the impact of single or combined vector control interventions. Methods build on Kipingu et al. (2025) <doi:10.1186/s12936-025-05454-y> and Johnson et al. (2015) <doi:10.1111/2041-210X.12306>.

Author(s)

Maintainer: Andrea M. Kipingu akipingu@ihi.or.tz (ORCID) (Environmental Health & Ecological Sciences Department, Ifakara Health Institute, P.O. Box 78 373, Dar es Salaam, Tanzania)

Authors:

• Mafalda Viana mafalda.viana@glasgow.ac.uk (ORCID) (School of Biodiversity, One Health and Veterinary Medicine, University of Glasgow, Graham Kerr Building, Glasgow, G12 8QQ, UK)

• Paul C.D. Johnson paul.johnson@glasgow.ac.uk (ORCID) (School of Biodiversity, One Health & Veterinary Medicine, University of Glasgow, Wolfson Link Building, Glasgow, G12 8QQ, UK)

• Samson S. Kiware skiware@ihi.or.tz (ORCID) (Environmental Health & Ecological Sciences Department, Ifakara Health Institute, P.O. Box 78 373, Dar es Salaam, Tanzania)

See Also

Useful links:

• https://github.com/akipingu/simpowa

• Report bugs at https://github.com/akipingu/simpowa/issues

Simulate Mosquito Count Data for Long-Term Semi-Field Experiment Testing Combined Interventions

Description

Generates simulated mosquito count data under a short-term semi-field experimental design with fixed effects for interventions 'intvn1' and 'intvn2', random effects for chambers and sampling time points variability, and a negative binomial distributed outcomes. Uses output from 'sim.scen.longsfe.comint()' to incorporate the table of experimental design scenarios.

Usage

sim.mosq.longsfe.comint(
  n.ch.per.trt,
  exp.length,
  sampl.freq = "weekly",
  lambda,
  intvn1.effect,
  intvn2.effect,
  ixn.effect,
  chamber.var,
  time.var,
  theta,
  use.random = "ALL"
)

Arguments

Value

A data frame with columns:

replicates

Replicate number within each treatment group (e.g., for a total of 2 chambers per treatment means 1, 2 replicates per treatment)

intvn1

Intervention status (e.g., 0 = control or no intervention 1 and 1 = there is intervention 1)

intvn2

Intervention status (e.g., 0 = control or no intervention 2 and 1 = there is intervention 2)

chamber

Unique chamber identifier as a factor (e.g., 0-0-1, 0-0-2 for the control chambers and 0-1-1, 1-0-1, 1-1-1 for intervention chambers)

time

Time sequence scaled between 0 and 1, it is calculate by the ratio of sampling points by their total number

timef

This is just the same as "time" but as a factor

ixn.intvn1.time

This is the interaction between an intvn1 and time, i.e., intvn1 x time

ixn.intvn2.time

This is the interaction between an intvn2 and time, i.e., intvn2 x time

ixn.intvn1.intvn2.time

This is the interaction between an intvn1, intvn2 and time, i.e., intvn1 x intvn2 x time

lin.pred.fixed

Linear predictor with fixed effects only

lin.pred.random

Linear predictor with random chamber effects

mosquito.count.fixed.exp

Simulted mosquito counts through exponetial function from fixed effects only (no sampling)

mosquito.count.fixed

Simulated mosquito counts through sampling distribution based on fixed effects only

mosquito.count.random

Simulated mosquito counts through sampling distribution accounting for random effects

Examples

sim.mosq.longsfe.comint(
  n.ch.per.trt = 2,
  exp.length = 90,
  sampl.freq = "monthly",
  lambda = 10,
  intvn1.effect = 0.7,
  intvn2.effect = 0.8,
  ixn.effect = 0.8,
  chamber.var = 0.1807,
  time.var = 0.2266,
  theta = 10,
  use.random = "ALL"
)

Simulate Mosquito Count Data for Long-Term Semi-Field Experiment Testing Single Intervention

Description

Generates simulated mosquito count data under a short-term semi-field experimental design with fixed effects for intervention ('intvn'), random effects for chambers and sampling time points variability, and a negative binomial distributed outcomes. Uses output from 'sim.scen.longsfe.sinint()' to incorporate the table of experimental design scenarios.

Usage

sim.mosq.longsfe.sinint(
  n.ch.per.trt,
  exp.length,
  sampl.freq = "weekly",
  lambda,
  intvn.effect,
  chamber.var,
  time.var,
  theta,
  use.random = "ALL"
)

Arguments

Value

A data frame with columns:

replicates

Replicate number within each treatment group (e.g., for a total of 2 chambers per treatment means 1, 2 replicates per treatment)

intvn

Intervention status (e.g., 0 = control or no intervention and 1 = there is intervention)

chamber

Unique chamber identifier as a factor (e.g., 0-1, 0-2 for the control chambers and 1-1, 1-2 for intervention chambers)

time

Time sequence scaled between 0 and 1, it is calculate by the ratio of sampling points by their total number

timef

This is just the same as "time" but as a factor

intvn.time

This is the interaction between an intervention and time, i.e., intvn x time

lin.pred.fixed

Linear predictor with fixed effects only

lin.pred.random

Linear predictor with random chamber effects

mosquito.count.fixed.exp

Simulted mosquito counts through exponetial function from fixed effects only (no sampling)

mosquito.count.fixed

Simulated mosquito counts through sampling distribution based on fixed effects only

mosquito.count.random

Simulated mosquito counts through sampling distribution accounting for random effects

Examples

sim.mosq.longsfe.sinint(
  n.ch.per.trt = 2,
  exp.length = 90,
  sampl.freq = "monthly",
  lambda = 10,
  intvn.effect = 0.8,
  chamber.var = 0.1807,
  time.var = 0.2266,
  theta = 10,
  use.random = "ALL"
)

Simulate Mosquito Count Data for Short-Term Semi-Field Experiment Testing Combined Interventions

Description

Generates simulated mosquito count data under a short-term semi-field experimental design with fixed effects for interventions, 'intvn1' and 'intvn2', random effects for chamber variability, and Poisson-distributed outcomes. Uses output from 'sim.scen.shortsfe.comint()' to incorporate the table of experimental design scenarios.

Usage

sim.mosq.shortsfe.comint(
  n.ch.per.trt,
  lambda,
  intvn1.effect,
  intvn2.effect,
  ixn.effect,
  chamber.var,
  use.random = "ALL"
)

Arguments

Value

A data frame with the following columns, depending on the use.random option:

replicates

Replicate number within each treatment group (e.g., for a total of 2 chambers per treatment means 1, 2 replicates per treatment)

intvn1

Intervention status (e.g., 0 = control or no intervention 1 and 1 = there is intervention 1)

intvn2

Intervention status (e.g., 0 = control or no intervention 2 and 1 = there is intervention 2)

ixn

Interventions interactions status (e.g., 0 = control or no intervention and 1 = there is interaction)

chamber

Unique chamber identifier as a factor (e.g., 0-0-1, 0-0-2 for the control chambers and 0-1-1, 1-0-1, 1-1-1 for intervention chambers)

lin.pred.fixed

Linear predictor with fixed effects only

lin.pred.random

Linear predictor with random chamber effects

mosquito.count.fixed.exp

Simulted mosquito counts through exponetial function from fixed effects only (no sampling)

mosquito.count.fixed

Simulated mosquito counts through sampling distribution based on fixed effects only

mosquito.count.random

Simulated mosquito counts through sampling distribution accounting for random effects

Examples

sim.mosq.shortsfe.comint(
  n.ch.per.trt = 2,
  lambda = 50,
  intvn1.effect = 0.7,
  intvn2.effect = 0.8,
  ixn.effect = 0.5,
  chamber.var = 0.1807,
  use.random = "ALL"
)

Simulate Mosquito Count Data for Short-Term Semi-Field Experiment Testing Single Intervention

Description

Generates simulated mosquito count data under a short-term semi-field experimental design with fixed effects for intervention ('intvn'), random effects for chamber variability, and Poisson-distributed outcomes. Uses output from 'sim.scen.shortsfe.sinint()' to incorporate the table of experimental design scenarios.

Usage

sim.mosq.shortsfe.sinint(
  n.ch.per.trt,
  lambda,
  intvn.effect,
  chamber.var,
  use.random = "ALL"
)

Arguments

Value

A data frame with the following columns, depending on the use.random option:

replicates

Replicate number within each treatment group (e.g., for a total of 2 chambers per treatment means 1, 2 replicates per treatment)

intvn

Intervention status (e.g., 0 = control or no intervention and 1 = there is intervention)

chamber

Unique chamber identifier as a factor (e.g., 0-1, 0-2 for the control chambers and 1-1, 1-2 for intervention chambers)

lin.pred.fixed

Linear predictor with fixed effects only

lin.pred.random

Linear predictor with random chamber effects

mosquito.count.fixed.exp

Simulted mosquito counts through exponetial function from fixed effects only (no sampling)

mosquito.count.fixed

Simulated mosquito counts through sampling distribution based on fixed effects only

mosquito.count.random

Simulated mosquito counts through sampling distribution accounting for random effects

Examples

sim.mosq.shortsfe.sinint(
  n.ch.per.trt = 2,
  lambda = 50,
  intvn.effect = 0.8,
  chamber.var = 0.1807,
  use.random = "ALL"
)

Plot Mosquito Counts for Long-Term Semi-Field Experiment Testing Combined Interventions

Description

Generates a time series plot of mosquito counts based on either fixed effects only, random effects, or fixed effects (no sampling distribution), using data simulated by 'sim.mosq.longsfe.comint'.

Usage

sim.plot.longsfe.comint(
  n.ch.per.trt,
  exp.length,
  sampl.freq = "weekly",
  lambda,
  intvn1.effect,
  intvn2.effect,
  ixn.effect,
  chamber.var,
  time.var,
  theta,
  use.random = TRUE
)

Arguments

Value

A 'ggplot' object showing mosquito counts over time by treatment group.

Examples

sim.plot.longsfe.comint(
  n.ch.per.trt = 2,
  exp.length = 90,
  sampl.freq = "monthly",
  lambda = 10,
  intvn1.effect = 0.7,
  intvn2.effect = 0.8,
  ixn.effect = 0.8,
  chamber.var = 0.1807,
  time.var = 0.2266,
  theta = 10,
  use.random = TRUE
)

Plot Mosquito Counts for Long-Term Semi-Field Experiment Testing Single Intervention

Description

Generates a time series plot of mosquito counts based on use.random with either fixed effects only, random effects, or fixed effects (no sampling distribution), using data simulated by 'sim.mosq.longsfe.sinint'.

Usage

sim.plot.longsfe.sinint(
  n.ch.per.trt,
  exp.length,
  sampl.freq = "weekly",
  lambda,
  intvn.effect,
  chamber.var,
  time.var,
  theta,
  use.random = TRUE
)

Arguments

Value

A 'ggplot' object showing mosquito counts over time by treatment group.

Examples

sim.plot.longsfe.sinint(
  n.ch.per.trt = 2,
  exp.length = 90,
  sampl.freq = "monthly",
  lambda = 10,
  intvn.effect = 0.8,
  chamber.var = 0.1807,
  time.var = 0.2266,
  theta = 10,
  use.random = TRUE
)

Plot Mosquito Counts for Short-Term Semi-Field Experiment Testing Combined Interventions

Description

Generates a boxplot of simulated mosquito counts grouped by intervention combinations. Uses output from 'sim.mosq.shortsfe.comint()' and overlays jittered points to show chamber-level variation.

Usage

sim.plot.shortsfe.comint(
  n.ch.per.trt,
  lambda,
  intvn1.effect,
  intvn2.effect,
  ixn.effect,
  chamber.var,
  use.random = TRUE,
  jitter = TRUE
)

Arguments

Value

A 'ggplot' object showing expected mosquito counts by treatment group.

Examples

sim.plot.shortsfe.comint(
  n.ch.per.trt = 2,
  lambda = 50,
  intvn1.effect = 0.7,
  intvn2.effect = 0.8,
  ixn.effect = 0.5,
  chamber.var = 0.1807,
  use.random = TRUE,
  jitter = TRUE
)

Plot Mosquito Counts from Short-Term Semi-Field Experiment Testing Single Intervention

Description

Generates a boxplot of simulated mosquito counts grouped by intervention status (Control vs Intervention). Uses output from 'sim.mosq.shortsfe.sinint()' and can overlay jittered points to show chamber-level variation.

Usage

sim.plot.shortsfe.sinint(
  n.ch.per.trt,
  lambda,
  intvn.effect,
  chamber.var,
  use.random = TRUE,
  jitter = TRUE
)

Arguments

Value

A 'ggplot' object showing expected mosquito counts by treatment group.

Examples

sim.plot.shortsfe.sinint(
  n.ch.per.trt = 2,
  lambda = 50,
  intvn.effect = 0.8,
  chamber.var = 0.1807,
  use.random = TRUE,
  jitter = TRUE
)

Estimate Empirical Power for Long-Term Semi-Field Experiment Testing Combined Interventions

Description

Runs repeated simulations and negative binomial GLMM fits to estimate empirical power as the proportion of simulations with p-values below 0.05. These p-values are generated using the function called 'sim.pval.longsfe.comint()'.

Usage

sim.power.longsfe.comint(
  n.ch.per.trt,
  exp.length,
  sampl.freq,
  lambda,
  intvn1.effect,
  intvn2.effect,
  ixn.effect,
  chamber.var,
  time.var,
  theta,
  nsim = 100,
  n.cores = 1,
  use.random = TRUE
)

Arguments

Value

A named numeric vector:

power

Estimated empirical power (rounded to two decimal places)

ci.lower

Lower bound of 95% confidence interval

ci.upper

Upper bound of 95% confidence interval

Note

Parallel execution is supported via 'n.cores', but examples default to 'n.cores = 1' for reproducibility and package checks.

Examples

# For a realistic SFE design, nsim should be at least 1000.


sim.power.longsfe.comint(
  n.ch.per.trt = 2,
  exp.length = 90,
  sampl.freq = "monthly",
  lambda = 10,
  intvn1.effect = 0.7,
  intvn2.effect = 0.8,
  ixn.effect = 0.8,
  chamber.var = 0.1807,
  time.var = 0.2266,
  theta = 10,
  nsim = 2,
  n.cores = 1,
  use.random = TRUE
)

Estimate Empirical Power for Long-Term Semi-Field Experiment Testing Single Intervention

Description

Runs repeated simulations and negative binomial GLMM fits to estimate empirical power as the proportion of simulations with p-values below 0.05. These p-values are generated using the function called 'sim.pval.longsfe.sinint()'.

Usage

sim.power.longsfe.sinint(
  n.ch.per.trt,
  exp.length,
  sampl.freq,
  lambda,
  intvn.effect,
  chamber.var,
  time.var,
  theta,
  nsim = 100,
  n.cores = 1,
  use.random = TRUE
)

Arguments

Value

A named numeric vector:

power

Estimated empirical power (rounded to two decimal places)

ci.lower

Lower bound of 95% confidence interval

ci.upper

Upper bound of 95% confidence interval

Note

Parallel execution is supported via 'n.cores', but examples default to 'n.cores = 1' for reproducibility and package checks.

Examples

# For a realistic SFE design, nsim should be at least 1000.

sim.power.longsfe.sinint(
  n.ch.per.trt = 2,
  exp.length = 90,
  sampl.freq = "monthly",
  lambda = 10,
  intvn.effect = 0.8,
  chamber.var = 0.1807,
  time.var = 0.2266,
  theta = 10,
  nsim = 2,
  n.cores = 1,
  use.random = TRUE
)

Estimate Empirical Power for Short-Term Semi-Field Experiment Testing Combined Intervention

Description

Runs repeated simulations and Poisson GLMM fits to estimate empirical power as the proportion of simulations with p-values below 0.05. These p-values are generated using the function called 'sim.pval.shortsfe.comint()'.

Usage

sim.power.shortsfe.comint(
  n.ch.per.trt,
  lambda,
  intvn1.effect,
  intvn2.effect,
  ixn.effect,
  chamber.var,
  nsim,
  n.cores = 1,
  use.random = TRUE
)

Arguments

Value

A named numeric vector:

power

Estimated empirical power (rounded to two decimal places)

ci.lower

Lower bound of 95% confidence interval

ci.upper

Upper bound of 95% confidence interval

Note

Parallel execution is supported via 'n.cores', but examples default to 'n.cores = 1' for reproducibility and package checks.

Examples

# For a realistic SFE design, nsim should be at least 1000.

sim.power.shortsfe.comint(
  n.ch.per.trt = 2,
  lambda = 50,
  intvn1.effect = 0.7,
  intvn2.effect = 0.8,
  ixn.effect = 0.5,
  chamber.var = 0.1807,
  nsim = 2,
  n.cores = 1,
  use.random = TRUE
)

Estimate Empirical Power for Short-Term Semi-Field Experiment Testing Single Intervention

Description

Runs repeated simulations and Poisson GLMM fits to estimate empirical power as the proportion of simulations with p-values below 0.05. These p-values are generated using the function called 'sim.pval.shortsfe.sinint()'.

Usage

sim.power.shortsfe.sinint(
  n.ch.per.trt,
  lambda,
  intvn.effect,
  chamber.var,
  nsim,
  n.cores = 1,
  use.random = TRUE
)

Arguments

Value

A named numeric vector:

power

Estimated empirical power (rounded to two decimal places)

ci.lower

Lower bound of 95% confidence interval

ci.upper

Upper bound of 95% confidence interval

Note

Parallel execution is supported via 'n.cores', but examples default to 'n.cores = 1' for reproducibility and package checks.

Examples

# For a realistic SFE design, nsim should be at least 1000.

sim.power.shortsfe.sinint(
  n.ch.per.trt = 2,
  lambda = 50,
  intvn.effect = 0.8,
  chamber.var = 0.1807,
  nsim = 2,
  n.cores = 1,
  use.random = TRUE
)

Extract p-value from Simulated GLMM for Long-Term Semi-Field Experiment Testing Combined Interventions

Description

Returns the p-value by fitting a negative binomial GLMM to simulated mosquito count data. Uses simulated mosquito counts data from 'sim.mosq.longsfe.comint()' and fits a negative binomial GLMM to extract the p-value associated with the intvn effect.

Usage

sim.pval.longsfe.comint(
  n.ch.per.trt,
  exp.length,
  sampl.freq = "weekly",
  lambda,
  intvn1.effect,
  intvn2.effect,
  ixn.effect,
  chamber.var,
  time.var,
  theta,
  use.random = TRUE
)

Arguments

Value

A named numeric vector:

pvalue

A p-value from likelihood ratio test comparing models with and without the interaction term

Examples

# For a realistic SFE design, nsim should be at least 1000
sim.pval.longsfe.comint(
  n.ch.per.trt = 2,
  exp.length = 90,
  sampl.freq = "monthly",
  lambda = 10,
  intvn1.effect = 0.7,
  intvn2.effect = 0.8,
  ixn.effect = 0.8,
  chamber.var = 0.1807,
  time.var = 0.2266,
  theta = 10,
  use.random = TRUE
)

Extract p-value from Simulated GLMM for Long-Term Semi-Field Experiment Testing Single Intervention

Description

Returns the p-value by fitting a negative binomial GLMM to simulated mosquito count data. Uses simulated mosquito counts data from 'sim.mosq.longsfe.sinint()' and fits a negative binomial GLMM to extract the p-value associated with the intvn effect.

Usage

sim.pval.longsfe.sinint(
  n.ch.per.trt,
  exp.length,
  sampl.freq = "weekly",
  lambda,
  intvn.effect,
  chamber.var,
  time.var,
  theta,
  use.random = TRUE
)

Arguments

Value

A named numeric vector:

pvalue

A p-value from likelihood ratio test comparing models with and without the interaction term

Examples

sim.pval.longsfe.sinint(
  n.ch.per.trt = 2,
  exp.length = 90,
  sampl.freq = "monthly",
  lambda = 10,
  intvn.effect = 0.8,
  chamber.var = 0.1807,
  time.var = 0.2266,
  theta = 10,
  use.random = TRUE
)

Extract p-value from Simulated GLMM for Short-Term Semi-Field Experiment Testing Combined Interventions

Description

Returns the p-value by fitting a Poisson GLMM to simulated mosquito count data. Uses simulated mosquito counts data from 'sim.mosq.shortsfe.comint()' and fits a Poisson GLMM to extract the p-value associated with the intvn effect.

Usage

sim.pval.shortsfe.comint(
  n.ch.per.trt,
  lambda,
  intvn1.effect,
  intvn2.effect,
  ixn.effect,
  chamber.var,
  use.random = TRUE
)

Arguments

Value

A named numeric vector:

pvalue

A p-value from likelihood ratio test comparing models with and without the interaction term

Examples

sim.pval.shortsfe.comint(
  n.ch.per.trt = 2,
  lambda = 50,
  intvn1.effect = 0.7,
  intvn2.effect = 0.8,
  ixn.effect = 0.5,
  chamber.var = 0.1807,
  use.random = TRUE
)

Extract p-values from Simulated GLMM for Short-Term Semi-Field Experiment Testing Single Intervention

Description

Returns the p-value by fitting a Poisson GLMM to simulated mosquito count data. Uses simulated mosquito counts data from 'sim.mosq.shortsfe.sinint()' and fits a Poisson GLMM to extract the p-value associated with the intvn effect.

Usage

sim.pval.shortsfe.sinint(
  n.ch.per.trt,
  lambda,
  intvn.effect,
  chamber.var,
  use.random = TRUE
)

Arguments

Value

A named numeric vector:

pvalue

A p-value for the intervention effect from the GLMM

Examples

sim.pval.shortsfe.sinint(
  n.ch.per.trt = 2,
  lambda = 50,
  intvn.effect = 0.8,
  chamber.var = 0.1807,
  use.random = TRUE
)

Simulate Design Scenarios for Long-Term Semi-Filed Experiment Testing Combined Interventions

Description

Constructs a data frame representing the factorial design of semi-field experiments testing combined interventions. Each chamber is uniquely identified and assigned treatments combinations, including interaction terms. The intervention here are shortly named as 'intvn1' and 'intvn2'.

Usage

sim.scen.longsfe.comint(n.ch.per.trt, exp.length, sampl.freq = "weekly")

Arguments

Value

A data frame with columns:

replicates

Replicate number within each treatment group (e.g., for a total of 2 chambers per treatment means 1, 2 replicates per treatment)

intvn1

Intervention status (e.g., 0 = control or no intervention 1 and 1 = there is intervention 1)

intvn2

Intervention status (e.g., 0 = control or no intervention 2 and 1 = there is intervention 2)

chamber

Unique chamber identifier as a factor (e.g., 0-0-1, 0-0-2 for the control chambers and 0-1-1, 1-0-1, 1-1-1 for intervention chambers)

time

Time sequence scaled between 0 and 1, it is calculate by the ratio of sampling points by their total number

timef

This is just the same as "time" but as a factor

ixn.intvn1.time

This is the interaction between an intvn1 and time, i.e., intvn1 x time

ixn.intvn2.time

This is the interaction between an intvn2 and time, i.e., intvn2 x time

ixn.intvn1.intvn2.time

This is the interaction between an intvn1, intvn2 and time, i.e., intvn1 x intvn2 x time

Examples

sim.scen.longsfe.comint(
  n.ch.per.trt = 2,
  exp.length = 90,
  sampl.freq = "monthly"
)

Simulate Design Scenarios for Long-Term Semi-Field Experiment Testing Single Intervention

Description

Constructs a design scenario table representing a short-term semi-field experiment with two treatment levels (or intervention status), i.e., control (no intervention) and intervention, and a specified number of chambers or compartment per treatment. Each chamber is uniquely identified by a treatment level and a replicate combination. An intervention here is shortly named as 'intvn'.

Usage

sim.scen.longsfe.sinint(n.ch.per.trt, exp.length, sampl.freq = "weekly")

Arguments

Value

A data frame with columns:

replicates

Replicate number within each treatment group (e.g., for a total of 2 chambers per treatment means 1, 2 replicates per treatment)

intvn

Intervention status (e.g., 0 = control or no intervention and 1 = there is intervention)

chamber

Unique chamber identifier as a factor (e.g., 0-1, 0-2 for the control chambers and 1-1, 1-2 for intervention chambers)

time

Time sequence scaled between 0 and 1, it is calculate by the ratio of sampling points by their total number

timef

This is just the same as "time" but as a factor

intvn.time

This is the interaction between an intervention and time, i.e., intvn x time

Examples

sim.scen.longsfe.sinint(
n.ch.per.trt = 2,
exp.length = 90,
sampl.freq = "monthly"
)

Simulate Design Scenarios for Short-Term Semi-Field Experiment Testing Combined Interventions

Description

Constructs a data frame representing the factorial design of semi-field experiments testing combined interventions. Each chamber is uniquely identified and assigned treatments combinations, including interaction terms. The intervention here are shortly named as 'intvn1' and 'intvn2'.

Usage

sim.scen.shortsfe.comint(n.ch.per.trt)

Arguments

Value

A data frame with columns:

replicates

Replicate number within each treatment group (e.g., for a total of 2 chambers per treatment means 1, 2 replicates per treatment)

intvn1

Intervention status (e.g., 0 = control or no intervention 1 and 1 = there is intervention 1)

intvn2

Intervention status (e.g., 0 = control or no intervention 2 and 1 = there is intervention 2)

ixn

Interventions interactions status (e.g., 0 = control or no intervention and 1 = there is interaction)

chamber

Unique chamber identifier as a factor (e.g., 0-0-1, 0-0-2 for the control chambers and 0-1-1, 1-0-1, 1-1-1 for intervention chambers)

Examples

sim.scen.shortsfe.comint(n.ch.per.trt = 2)

Simulate Design Scenarios for Short-Term Semi-Field Experiment Testing Single Intervention

Description

Constructs a design scenario table representing a short-term semi-field experiment with two treatment levels (or intervention status), i.e., control (no intervention) and intervention, and a specified number of chambers or compartment per treatment. Each chamber is uniquely identified by a treatment level and a replicate combination. An intervention here is shortly named as 'intvn'.

Usage

sim.scen.shortsfe.sinint(n.ch.per.trt)

Arguments

Details

This function considers a single design parameter, which is the total number of chambers or compartments per treatment ('n.ch.per.trt'), e.g., 4. The 'n.ch.per.trt' indicates how many chambers known as 'replicates' are present per treatment.

Value

A data frame with columns:

replicates

Replicate number within each treatment group (e.g., for a total of 2 chambers per treatment means 1, 2 replicates per treatment)

intvn

Intervention status (e.g., 0 = control or no intervention and 1 = there is intervention)

chamber

Unique chamber identifier as a factor (e.g., 0-1, 0-2 for the control chambers and 1-1, 1-2 for intervention chambers)

Examples

sim.scen.shortsfe.sinint(n.ch.per.trt = 2)