Title: Log File Analysis in International Large-Scale Assessments
Version: 1.0.1
Date: 2022-10-24
Description: Enables users to handle the dataset cleaning for conducting specific analyses with the log files from two international educational assessments: the Programme for International Student Assessment (PISA, https://www.oecd.org/pisa/) and the Programme for the International Assessment of Adult Competencies (PIAAC, https://www.oecd.org/skills/piaac/). An illustration of the analyses can be found on the LOGAN Shiny app (https://loganpackage.shinyapps.io/shiny/) on your browser.
BugReports: https://github.com/derecost/LOGAN/issues
License: GPL-3
Encoding: UTF-8
LazyData: true
Depends: R (≥ 3.5)
Imports: pander (≥ 0.6.1), psych (≥ 1.7.8), foreign (≥ 0.8-69), dplyr, ggplot2, rlang, stringr, magrittr, modules, methods
Suggests: knitr, rmarkdown
RoxygenNote: 7.2.1
NeedsCompilation: no
Packaged: 2022-10-25 04:12:35 UTC; waldir
Author: Denise Reis Costa [aut], Waldir Leoncio [aut, cre]
Maintainer: Waldir Leoncio <w.l.netto@medisin.uio.no>
Repository: CRAN
Date/Publication: 2022-10-25 08:47:56 UTC

Plot: Boxplot PV1Math by Strategy var

Description

Plot: Boxplot PV1Math by Strategy var

Usage

BoxplotStrategybyPerformance(
  data,
  strategy.var,
  performance.test,
  ylab.text,
  xlab.text
)

Arguments

data

data

strategy.var

strategy.var

performance.test

performance.test

ylab.text

ylab.text

xlab.text

xlab.text

Clean events

Description

This function allows you to clean events in the 'event.type' variable

Usage

CleanActions(data, event.type, clear.events)

Arguments

data

A matrix or data.frame where the 'event.type' variable is

event.type

a vector with concatenate events. See ConcatActions function.

clear.events

a vector where all the events to be cleaned are listed. Each element of this vector needs to be of a "event"="" type.

Value

This function returns a data.frame with the "new.event.type" variable that cleaned events from the "event.type" variable.

Examples

# Data preparation
df <- cp025q01
df$id <- paste(df[, 1], df[, 2], df[, 3], sep = "-")
df <- m0$TrimVar(df, c("event", "event_type", "diag_state"))
df <- m0$ConcatActions(df, c(rlang::quo(event), rlang::quo(event_type)))

# Function demonstration
df.clean <- m0$CleanActions(df, event_type, c("ACER_EVENT_" = ""))
table(df$event.type)
table(df.clean$new.event.type) # cleaned version

Concatenate events

Description

This function allows you to concatenate event actions from diferent variables in a unique vector.

Usage

ConcatActions(data, concat.events)

Arguments

data

A matrix or data.frame where the concatenated events are

concat.events

a vector where all the events are listed. Each element of this vector needs to be of a quo() type.

Details

The output dataset will be identical to the input dataset, except for the addition of one column in the end, called "event.type". Each row of event.type contains the values of concat.events of all the rows.

Value

This function returns a data.frame with the concatenated events in the 'event.type' variable.

Examples

# Data preparation
df <- cp025q01
df$id <- paste(df[, 1], df[, 2], df[, 3], sep = "-")
df <- m0$TrimVar(df, c("event", "event_type", "diag_state"))

# Function demonstration
df.conc <- m0$ConcatActions(df, c(rlang::quo(event), rlang::quo(event_type)))
names(df)
names(df.conc) # notice the extra variable in the end
table(df.conc$event.type)

Wide format dataset with the sequence of actions by ID

Description

This is a function that translates a long to wide format dataset.

Usage

DataActionsbyID(data, id.var, event.var, name.var.action)

Arguments

data

A matrix or data.frame where the 'event.type' variable is

id.var

a vector with the individuals identification. It is a quo() type.

event.var

a vector with the cleaned concatenate events. See CleanActions function.

name.var.action

A character string that will name the new variable of events

Value

This function returns a data.frame with the only one entry by individual identification and a new 'action.var' variable.

Examples

# Data preparation
df <- cp025q01
df$id <- paste(df[, 1], df[, 2], df[, 3], sep = "-")
df <- m0$TrimVar(df, c("event", "event_type", "diag_state"))
df <- m0$ConcatActions(df, c(rlang::quo(event), rlang::quo(event_type)))
df <- m0$CleanActions(df, event.type, c("ACER_EVENT_" = ""))

# Function demonstration
m0$DataActionsbyID(df, id, new.event.type, "actions")

Data: Percentage in arcsine values x PISA scores by Country

Description

This is a function that calculates the percentage in arcsine and plots it against the PISA scores

Usage

DataArcSinebyPerformance(data, strategy.var, performance.test, country.id)

Arguments

data

A matrix or data.frame where the 'strategy.var' and performance variables are

strategy.var

A string with the name of the strategy variable. It is "quo()" type.

performance.test

A string with the name of the test performance variable. It is "quo()" type.

country.id

A string with the name of the countries variable. It is "quo()" type.

Value

This function returns a data frame and a plot

Report: Descriptive statistics by strategy

Description

This is a function that reports a descriptive analysis of the strategy and students performance

Usage

DescriptiveStrategy(
  data,
  strategy.var,
  performance.item,
  performance.test,
  PartialCredit = FALSE
)

Arguments

data

A matrix or data.frame where the 'strategy.var' and performance variables are

strategy.var

A character string with the name of the strategy variable

performance.item

A character string with the name of the item performance variable

performance.test

A character string with the name of the test performance variable

PartialCredit

Logical. It can be used when the item is partial credit score.

Value

This function returns a report with a descriptive analysis of the strategy and students performance

Examples

m2$DescriptiveStrategy(cp025q01.treated, "votat", "CP025Q01", "PV1CPRO")

Frequency of specifics events in a variable of Actions - Summary

Description

This is a function that locates specific events (using the actions.search argument) and create new variables associate with this strategy.

Usage

FreqActionsSummary(data, freqact.var, var)

Arguments

data

A matrix or data.frame where the 'action.var' variable is

freqact.var

freqact.var

var

var

Value

This function returns a data.frame with the frequency of each specific events from the actions.search argument and "Freq.Actions.Search" summary.

Read SPSS process data

Description

This is a simple function that, by default, reads an SPSS data file and save it as a data frame. It is essentially a wrapper for foreign::read.spss with arguments common to log file datasets.

Usage

ImportSPSS(filename)

Arguments

filename

character string: the name of the file or URL to read.

Value

This function returns a data frame.

LOGAN: Log File Analysis in International Large-scale Assessments

Description

This package enables users to handle the dataset cleaning for conducting specific analyses with the log files from two international educational assessments: the Programme for International Student Assessment (PISA, <http://www.oecd.org/pisa/>) and the Programme for the International Assessment of Adult Competencies (PIAAC, <http://www.oecd.org/skills/piaac/>). An illustration of the analyses can be found on the LOGAN Shiny app (<https://loganpackage.shinyapps.io/shiny/>) on your browser.

LOGAN functions

The LOGAN functions The LOGAN functions are organized in modules, so to call a function you must prefix it with, e.g., 'm0$', where "m0" is the module to which a certain function pertains.

What follows is a list of Functions organized per module:

Module 0:

Module 1:

Module 2:

Author(s)

Time var as a numeric vector

Description

This is a function that transforms a factor var time in numeric.

Usage

NumericTimeVar(data, vector.time)

Arguments

data

A matrix or data.frame

vector.time

variable containing the time

Value

This function returns a data.frame with the number of students and number de actions (min-max) aggregated by a specific variable.

Examples

vector.time <- c("CP025Q01.END", "CP025Q01.START")
m1$NumericTimeVar(cp025q01.treated, vector.time)

Check response time by var

Description

This is a function that reports the number of students and number of actions (min-max) aggregated by a specific variable.

Usage

PlotStrategybyCatPerformance(data, strategy.var, categ.var, namexlab, nameylab)

Arguments

data

A matrix or data.frame

strategy.var

strategy variable

categ.var

categorizing variable

namexlab

name of the variable in the x-axis

nameylab

name of the variable in the y-axis

Value

This function returns a data.frame with the number of students and number de actions (min-max) aggregated by a specific variable.

Examples

# Data preparation
df <- cp025q01.treated
df$categ <- cut(df$PV1CPRO, c(0, 423, 488, 553, 900))
df.dataplot <- df[, c("top", "categ")]
df.dataplot[, 1] <- as.factor(df.dataplot[, 1])
df.dataplot[, 2] <- as.factor(df.dataplot[, 2])

# Function demonstration
m2$PlotStrategybyCatPerformance(
  df.dataplot, top, categ,
  "Proficiency levels", "Percentage"
)

Check response time by var

Description

This is a function that reports the number of students and number de actions (min-max) aggregated by a specific variable.

Usage

PlotTimeonTaskbyVar(
  data,
  tot.var,
  performance.item,
  namexlab,
  nameylab = "Density"
)

Arguments

data

A matrix or data.frame

tot.var

a vector with the total time. It is a quo() type.

performance.item

name of the item variable

namexlab

name of the plot's x-axis

nameylab

name of the plot's y-axis. Defaults to "Density"

Value

This function returns a data.frame with the number of students and number de actions (min-max) aggregated by a specific variable.

Examples

m1$PlotTimeonTaskbyVar(cp025q01.treated, "CP025Q01.TOT", "CP025Q01",
  namexlab = "Time on task (minutes)"
)

Check number of students and actions by var

Description

This is a function that reports the number of students and number de actions (min-max) aggregated by a specific variable.

Usage

RangeNumberActionsbyVar(data, id.var, var.group, save.table = TRUE)

Arguments

data

A matrix or data.frame

id.var

a vector with the individuals identification. It is a quo() type.

var.group

a vector with the group variable. It is a quo() type.

save.table

if TRUE, will save the table generated as an object of class data.frame. Otherwise, will print the table in pandoc format, but the object will not be saved (even if the user assigns it to an object)

Value

This function returns a data.frame with the number of students and number de actions (min-max) aggregated by a specific variable.

Examples

m0$RangeNumberActionsbyVar(cp025q01.treated, NewID, CNT, save.table = FALSE)

Check response time by var

Description

This is a function that reports the number of students and number de actions (min-max) aggregated by a specific variable.

Usage

RangeTimeonTaskbyVar(data, tot.var, var.group)

Arguments

data

A matrix or data.frame

tot.var

a vector with the total time. It is a quo() type.

var.group

a vector with the group variable. It is a quo() type.

Value

This function returns a data.frame with the number of students and number de actions (min-max) aggregated by a specific variable.

Plot: Percentage in arcsine values x PISA scores by Country

Description

This is a function that calculates the percentage in arcsine and plots it against the PISA scores

Usage

ScatterPlotbyPerformance(
  data,
  strategy.summary,
  performance.mean,
  country.id,
  ylab.text,
  xlab.text,
  ylim.vector,
  xlim.vector
)

Arguments

data

A matrix or data.frame where the 'strategy.var' and performance variables are

strategy.summary

strategy.summary

performance.mean

performance.mean

country.id

A string with the name of the countries variable. It is "quo()" type.

ylab.text

A character string giving the text of the y-axis in the plot

xlab.text

A character string giving the text of the x-axis in the plot

ylim.vector

A numeric vector with the limits of the y-axis in the plot

xlim.vector

A numeric vector with the limits of the x-axis in the plot

Value

This function returns a data frame and a plot

Summary of time on task by var

Description

This is a function that reports the number of students and a summary of time on task aggregated by a specific variable.

Usage

SummaryTOTbyVar(data, tot.var, performance.item, na.rm = FALSE)

Arguments

data

A matrix or data.frame

tot.var

a vector with the time on task.

performance.item

a vector with the group variable. It is a quo() type.

na.rm

remove missing data in 'performance.item'? Default is 'FALSE'

Value

This function returns a data.frame with the number of students and number de actions (min-max) aggregated by a specific variable.

Examples

m1$SummaryTOTbyVar(cp025q01.treated, "CP025Q01.TOT", "CP025Q01", TRUE)

Time on task variable

Description

This is a function that reports the number of students and a summary of time on task aggregated by a specific variable.

Usage

TOTVar(data, starttime.vec, endtime.vec, divBy = NA, tot.var)

Arguments

data

A matrix or data.frame

starttime.vec

a vector with the individuals' identifications. It is a quo() type.

endtime.vec

a vector with the group variable. It is a quo() type.

divBy

a vector with the group variable. It is a quo() type.

tot.var

string containing the name of the output variable

Value

This function returns a data.frame with the number of students and number de actions (min-max) aggregated by a specific variable.

Examples

m1$TOTVar(cp025q01.treated, "CP025Q01.START", "CP025Q01.END",
  divBy = 60,
  tot.var = "CP025Q01.TOT"
)

Trim variables

Description

TrimVar() is a function that allows you to remove whitespace inside the strings of a vector.

Usage

TrimVar(data, trim.vector)

Arguments

data

dataset

trim.vector

vector of variables on the dataset to be trimmed

Value

This function returns a vector removing trailing and leading spaces inside the original vector.

Examples

head(m0$TrimVar(cp025q01, "event"))

Identify the position of specific events in a variable of Actions

Description

This is a function that locates specific events (using the actions.search argument) and create new variables associate with this strategy.

Usage

VarActionPosition(data, action.var, actions.search)

Arguments

data

A matrix or data.frame where the 'action.var' variable is

action.var

a vector with actions. See DataActionsbyID function.

actions.search

A character vector with the actions to be searched.

Value

This function returns a data.frame with the frequency of each specific events from the actions.search argument and "Freq.Actions.Search" summary.

Frequency of specifics events in a variable of Actions

Description

This is a function that locates specific events (using the actions.search argument) and create new variables associate with this strategy.

Usage

VarActionSearch(data, action.var, actions.search)

Arguments

data

A matrix or data.frame where the 'action.var' variable is

action.var

a vector with actions. See DataActionsbyID function.

actions.search

A character vector with the actions to be searched.

Value

This function returns a data.frame with the frequency of each specific events from the actions.search argument and "Freq.Actions.Search" summary.

Examples

# Counting the instances of top_setting == 1
df <- m2$VarActionSearch(cp025q01.treated, "CP025Q01.ACTIONS", "1_apply")
table(df$freq.1_apply) # checking results

Extracting the start or end time

Description

Extracting the start or end time

Usage

VarTimebyID(data, id.var, time.var, event.var, name.var.time, new.name)

Arguments

data

data frame

id.var

vector of unique identification

time.var

vector with the time variable

event.var

vector with the events

name.var.time

name of the time string to filter (ex.: "START_ITEM" or "END_ITEM")

new.name

name of the output variable

Value

a data frame with 'time' replaced with 'new.name'. The variable 'event.var' is dropped.

Examples

# Data preparation
df <- cp025q01
df$id <- paste(df[, 1], df[, 2], df[, 3], sep = "-")
df <- m0$TrimVar(df, c("event", "event_type", "diag_state"))
df <- m0$ConcatActions(df, c(rlang::quo(event), rlang::quo(event_type)))
df <- m0$CleanActions(df, event.type, c("ACER_EVENT_" = ""))

# Function demonstration
m1$VarTimebyID(df, id, time, new.event.type, "START_ITEM", "start")

Log file for PISA 2012, CP025, Q01 (selected countries)

Description

Log file for PISA 2012, CP025, Q01 (selected countries)

Treated log file and microdata for PISA 2012, CP025, Q01 (selected countries)

Description

Treated log file and microdata for PISA 2012, CP025, Q01 (selected countries)

Module 0: Data preparation

Description

Module 0: Data preparation

Usage

m0

Format

An object of class module (inherits from list) of length 6.

Details

This module contains the following functions, which should be called by issuing "m0$<function_name>()": CleanActions, ConcatActions, DataActionsbyID, ImportSPSS, RangeNumberActionsbyVar, TrimVar

Module 1: Time

Description

Module 1: Time

Usage

m1

Format

An object of class module (inherits from list) of length 5.

Details

This module contains the following functions, which should be called by issuing "m1$<function_name>()": NumericTimeVar, PlotTimeonTaskbyVar, SummaryTOTbyVar, TOTVar, VarTimebyID

Module 2: Actions (cognitive related)

Description

Module 2: Actions (cognitive related)

Usage

m2

Format

An object of class module (inherits from list) of length 3.

Details

This module contains the following functions, which should be called by issuing "m2$<function_name>()": DescriptiveStrategy, PlotStrategybyCatPerformance, VarActionSearch.

Microdata for PISA 2012 (selected countries)

Description

Microdata for PISA 2012 (selected countries)

mirror server hosted at Truenetwork, Russian Federation.