README.md

In debruine/pipeline: Machine-Readable Studies

pipeline

The goal of pipeline is to generate and process machine-readable study descriptions.

Installation

You can install the released version of pipeline from GitHub with:

devtools::install_github("scienceverse/pipeline")

Example

Get the file paths to the json file and data. Load the study parameters and data with the pipeline() function; this will run the analyses.

library(pipeline)

# find file paths to data included with this package installation
iat_json_path <- system.file("extdata", "iat.json", package = "pipeline")
iat_data_path <- system.file("extdata", "iat.csv", package = "pipeline")

iat_study <- pipeline(iat_json_path, iat_data_path)

Generate a report. Set results='asis' in the code chunk.

summary(iat_study, "hypotheses", "analyses", "results")

Hypotheses

Hypothesis 1

Mean RT will be significantly slower in the incongruent condition compared to the congruent condition.

• Criterion 1 is confirmed if analysis yields p.value < 0.05
• Criterion 2 is confirmed if analysis yields estimate > 0

If all criteria are met, this hypothesis is supported.

Hypothesis 2

Reaction times for congruent and incongruent trials will be signifiantly and positively correlated.

• Criterion 1 is confirmed if analysis yields p.value < 0.05
• Criterion 2 is confirmed if analysis yields estimate > 0

If all criteria are met, this hypothesis is supported.

Analyses

H1_ttest

We will run t.test(x = .data$incongruent, y = .data$congruent, alternative = two.sided, paired = TRUE, var.equal = FALSE, conf.level = 0.95)

H2_cor

We will run cor.test(x = .data$incongruent, y = .data$congruent, alternative = two.sided, paired = TRUE, var.equal = FALSE, conf.level = 0.95)

Results

Hypothesis 1

Mean RT will be significantly slower in the incongruent condition compared to the congruent condition.

• Criterion 1 was p.value < 0.05 in analysis H1_ttest. The result was p.value = 0
• Criterion 2 was estimate > 0 in analysis H1_ttest. The result was estimate = 281.067

Conclusion: Congratulations! All criteria were met, this hypothesis was supported.

Hypothesis 2

Mean RT will be significantly slower in the incongruent condition compared to the congruent condition.

• Criterion 1 was p.value < 0.05 in analysis H2_cor. The result was p.value = 0.006
• Criterion 2 was estimate > 0 in analysis H2_cor. The result was estimate = 0.384

Conclusion: Congratulations! All criteria were met, this hypothesis was supported.

debruine/pipeline documentation built on May 8, 2019, 8:59 a.m.