Convenience functions to develop psychometric tests

testing is a package for the R statistical environment providing functions to develop psychometric tests, mainly leveraging on the Classical Test Theory.

The package can be installed by connecting to this repository by using the package devtools, which must be previously installed. To install testing you can run the command:

devtools::install_github("DavideMassidda/testing")

testing not replaces other well-known packages as psych (on which it depends), but should be considered as their integration.

The package stems from the author's need to collect into a single repository the most commonly R functions used in his daily practice. These functions are therefore rather heterogeneous, but they share the usefulness in some steps of psychometric test building. Some functions of testing are present in other forms in different R packages, when not in base R. In these cases, testing offers a different output format optimized for application in psychometric data analysis. Sometimes, the functions are implemented following algorithmic peculiarities that make them only apparently similar to other sources.

Born from a messy collection of commands scattered within a multitude of scripts, over time the package has been refined, making a great effort to optimize the functions and standardize the style. During the development, some functions were refined, and others were deprecated and not replaced. If you search old versions including functions no longer available, you can take a look of old sources of the package.

Package overview

Scoring

• reverse Score reversing.
• normalize (or normalise) Vector rescaling.
• stdscore From raw scores to standard scores.
• rawscore From standard scores to raw scores.
• percrank From observed scores to percentile ranks.
• raw2std Conversion of raw scores from a norm table.
• std2perc From standard scores to percentiles.
• perc2std From percentiles to standard scores.

Rounding

• integer_round Conversion from decimal to integer.
• decimal_floor Round a decimal number to the nearest floor for a given decimal position.
• decimal_ceiling Round a decimal number to the nearest ceiling for a given decimal position.
• integer_floor Round a decimal number to the nearest integer floor.
• integer_ceiling Round a decimal number to the nearest integer ceiling.

Norming

• score_rollup Creation of intervals of scores.
• score_explode Explosion of intervals of scores.
• score_implode Implosion of intervals of scores.
• is_continuous Check for continuous vectors.
• is_monotonic Check for monotonic vectors.
• roc_table Search a score cut-off by using a ROC analysis.

Validating

• se_measure Standard error of measurement.
• kr20 Kuder–Richardson formula 20.
• cronbach_alpha Cronbach's alpha.
• cronbach_strata Stratified Cronbach's alpha.
• average_reliability Average reliability coefficient.
• splithalf_reliability Split-half reliability.
• drop_item Internal consistency reliability.

Describing

• entropy Entropy index.
• integer_counts Count integer values.

Aging

• age_completed Age Calculation from dates.
• age_numeric Age Conversion: from character to numeric.
• age_character Age Conversion: from numeric to character.
• age_segment Grouping by age segmentation.

Datasets

• drive Secure drive project.
• abilityTest Ability test.
• normTab Normative table.
• testDates Test dates.

Documentation pages (Italian only)

▶️ Standardizzazione dei punteggi

▶️ Tabulazione di riferimenti normativi

▶️ Calcolo e conversione dell’età