Package 'psycCleaning'

Center with respect to grand mean

Description

This function will compute grand-mean-centered scores.

Usage

center_grand_mean(data, cols, keep_original = TRUE)

Arguments

data	A data.frame or a data.frame extension (e.g. a tibble).
cols	Columns that need to be centered. See 'dplyr::dplyr_tidy_select' for available options.
keep_original	default is 'FALSE'. Set to 'TRUE' to keep original columns

Value

An object of the same type as .data. The output has the following properties: 1. Columns from .data will be preserved 2. Columns with scores that are grand-mean-centered.

Examples

center_grand_mean(iris,where(is.numeric))

---

Center with respect to group mean

Description

This function will compute group-mean-centered scores.

Usage

center_group_mean(data, cols, group, keep_original = TRUE)

Arguments

data	A data.frame or a data.frame extension (e.g. a tibble).
cols	Columns that need to be centered. See 'dplyr::dplyr_tidy_select' for available options.
group	character. grouping variable
keep_original	default is 'TRUE'. Set to 'FALSE' to remove original columns

Value

An object of the same type as .data. The output has the following properties: 1. Columns from .data will be preserved 2. Columns with scores that are group-mean centered

Examples

center_group_mean(iris,where(is.numeric), group = Species)

---

Centering for multilevel analyses

Description

This function will group mean centered the scores at the level 1 and create a mean score for each group at L2.

Usage

center_mlm(data, cols, group, keep_original = TRUE)

Arguments

data	A data.frame or a data.frame extension (e.g. a tibble).
cols	Columns that need to be centered. See 'dplyr::dplyr_tidy_select' for available options.
group	the grouping variable. Must be character.
keep_original	default is 'TRUE'. Set to 'FALSE' to remove original columns

Value

An object of the same type as .data. The output has the following properties: 1. Columns from .data will be preserved 2. Columns with L1 scores that are group-mean centered. 3. Columns with L2 aggregated means.

Examples

center_mlm(iris,dplyr::ends_with('Length'),group = 'Species')

---

Composite column

Description

The function will perform a row-wise aggregation which then divided by the total number of columns.

Usage

composite_score(
  data,
  cols = dplyr::everything(),
  na.rm = FALSE,
  composite_col_name = "composited_column"
)

Arguments

data	A data.frame or a data.frame extension (e.g. a tibble).
cols	Columns that need to be composited See 'dplyr::dplyr_tidy_select' for available options.
na.rm	Ignore NA. The default is 'FALSE'. If set to 'TRUE', the composite score will be 'NA' if there is one or more 'NA' in any of the columns.
composite_col_name	Name for the new composited columns. Default is 'composite_column'.

Value

An object of the same type as .data. The output has the following properties: 1. Columns from .data will be preserved. 2. Columns with composited scores.

Examples

test_df = data.frame(col1 = c(1,2,3,4),col2 = c(1,2,3,4), col3 = c(1,2,NA,4))
composite_df = composite_score(data = test_df)

---

Dummy Coding

Description

Create dummy-coded columns, supporting tidyselect syntax to process multiple columns simultaneously.

Usage

dummy_coding(data, cols)

Arguments

data	data.frame object
cols	Columns that need to be dummy-coded See 'dplyr::dplyr_tidy_select' for available options.

Value

An object of the same type as .data. The output has the following properties: 1. Columns from .data will be preserved. 2. Columns that are dummy-coded.

Examples

dummy_coding(iris,Species)

---

Effect Coding

Description

Create effect-coded columns, supporting tidyselect syntax to process multiple columns simultaneously.

Usage

effect_coding(data, cols, factor = FALSE, ref_group = NULL)

Arguments

data	A data.frame or a data.frame extension (e.g. a tibble).
cols	Columns that need to be effect-coded. See 'dplyr::dplyr_tidy_select' for available options.
factor	The default is 'FALSE'. If factor is set to 'TRUE', this function returns a tibble with effect-coded factors. If factor is set to 'FALSE', this function returns a tibble with effect-coded columns.
ref_group	Reference group. Optional.

Value

An object of the same type as .data. The output has the following properties: 1. Columns from .data will be preserved. 2. Columns that are effect-coded.

Examples

effect_coding(iris,Species)

---

Listwise deletion

Description

Perform listwise deletion (the entire rows is disregarded if the row has one 'NA' value)

Usage

listwise_deletion(data, cols = dplyr::everything())

Arguments

data	A data.frame or a data.frame extension (e.g. a tibble).
cols	Columns that need to use listwise deletion. See 'dplyr::dplyr_tidy_select' for available options.

Value

An object of the same type as .data with rows revmoed if the row has one 'NA' value

Examples

test_df = data.frame(col1 = c(1,2,3),col2 = c(1,NA,3),col3 = c(1,2,NA))
listwise_deletion(test_df,col1:col2) # you can see that the row with NA in col3 is not deleted

---

mlbook_data

Description

Classic data-set from Snijders, Tom A.B., and Bosker, Roel J. Multilevel Analysis: An Introduction to Basic and Advanced Multilevel Modeling, second edition.

Usage

mlbook_data

Format

A data frame with 3758 rows and 34 variables:

schoolnr

School ID

pupilNR_new

Student Identifier (Level 1 units)

langPOST

Student language score

ses

Student socioeconomic score, grand-mean centered (in points, M = 0))

IQ_verb

Student verbal IQ, grand-mean centered (in points, M = 0)

sex

Student binary gender, 1 = female, 0 = not female

Minority

Student minority status, 1 = minoritized, 0 = not minoritized

denomina

School-level religious denominations, 5 categories

female_dum

Dummy coded sex

female_eff

Effect-coded sex

female_CMC

Group-mean-centered of female_eff

fempct_agg

Aggregated mean female_dum for each school

Zfempct_agg

Z-scored aggregated mean female_dum for each school

ses_CMC

Group-mean-centered SES

Zses_CMC

Z-scored group-mean-centered SES

ses_agg

Aggregated mean SES for each school

Zses_agg

Z-scored aggregated mean SES for each school

Source

https://www.stats.ox.ac.uk/~snijders/mlbook.htm

---

Recode values of a data frame

Description

Recode values of a data frame

Usage

recode_item(data, cols, code_from = NULL, code_to = NULL, retain_code = NULL)

Arguments

data	A data.frame or a data.frame extension (e.g. a tibble).
cols	Columns that need to be recoded. See 'dplyr::dplyr_tidy_select' for available options.
code_from	vector. the order must match with vector for 'code_to'
code_to	vector. the order must match with vector for 'code_from'
retain_code	vector. Specify the values to be retain

Value

An object of the same type as .data. The output has the following properties: 1. Columns except the recoded columns from .data will be preserved 2. Recoded columns

Examples

pre_recoded_df = tibble::tibble(x1 = 1:5, x2 = 5:1)
recoded_df = recode_item(pre_recoded_df, cols = dplyr::contains('x'),
                        code_from = 1:5,
                        code_to = 5:1)

---

Count the number of missing values

Description

It counts the number of missing (i.e.,'NA') values in each column.

Usage

summarize_missing_values(
  data,
  cols = dplyr::everything(),
  group = NULL,
  verbose = TRUE,
  return_result = FALSE
)

Arguments

data	A data.frame or a data.frame extension (e.g. a tibble).
cols	Columns that need to be checked for missing values. See 'dplyr::dplyr_tidy_select' for available options.
group	character. count missing values by group.
verbose	default is 'TRUE'. Print the missing value data frame
return_result	default is 'FALSE'. Return 'data_frame' if set to yes

Value

An object of the same type as .data. that specified the number of NA values of the columns (only when 'return_result = TRUE')

Examples

df1 = data.frame(col1 = c(1,2,3),col2 = c(1,NA,3),col3 = c(1,2,NA))
summarize_missing_values(df1,everything())

---

Grand mean z-score

Description

This function will compute z-scores with respect to the grand mean.

Usage

z_scored_grand_mean(data, cols, keep_original = TRUE)

Arguments

data	A data.frame or a data.frame extension (e.g. a tibble).
cols	Columns that need to be centered. See 'dplyr::dplyr_tidy_select' for available options.
keep_original	default is 'FALSE'. Set to 'TRUE' to keep original columns

Value

An object of the same type as .data. The output has the following properties: 1. Columns from .data will be preserved 2. Columns with scores that are z-scored

Examples

z_scored_grand_mean(iris,where(is.numeric))

---

Z scored with with respect to the group mean

Description

This function will compute group-mean-centered scores, and then z-scored the group-mean-centered scores with respect to the grand mean.

Usage

z_scored_group_mean(data, cols, group, keep_original = TRUE)

Arguments

data	A data.frame or a data.frame extension (e.g. a tibble).
cols	Columns that need to be centered. See 'dplyr::dplyr_tidy_select' for available options.
group	the grouping variable. If you need to pass multiple group variables, try to use quos(). Passing multiple group variables is not tested.
keep_original	default is 'FALSE'. Set to 'TRUE' to keep original columns

Value

return a dataframe with a group-mean centered columns that are z-scored with respect to the grand mean.

Examples

z_scored_group_mean(iris, dplyr::ends_with("Petal.Width"), "Species")

---

Z-scored for multilevel analyses

Description

This function will group mean centered the scores at the level 1 and create an aggregated mean score for each group at L2. After that, the group-mean-centered L1 scores and mean L2 scores will be z-scored with respect to the grand mean. Please see 'center_mlm' if you want to use the version without the z-scoring.

Usage

z_scored_mlm(data, cols, group, keep_original = TRUE)

Arguments

data	A data.frame or a data.frame extension (e.g. a tibble).
cols	Columns that need to be centered. See 'dplyr::dplyr_tidy_select' for available options.
group	The grouping/cluster variable.
keep_original	default is 'TRUE'. Set to 'FALSE' to remove original columns

Value

An object of the same type as .data. The output has the following properties: 1. Columns from .data will be preserved 2. Columns with L1 scores that are group-mean centered then grand-mean z-scored. 3. Columns with L2 aggregated means that are z-scored

Examples

z_scored_mlm(iris,dplyr::ends_with('Length'),group = 'Species')

---

Z-scored for multilevel analyses

Description

This is a specialized function for mean centering categorical variables. There are two cases where this function should be used instead of the generic 'center_mlm'. 1. This function should be used when you need group mean centering for non-dummy-coded variables at L1. Variables at L2 are always dummy-coded as they represent the percentage of subjects in that group. 2. This function should be used whenever you want to z-score the aggregated L2 means

Usage

z_scored_mlm_categorical(
  data,
  cols,
  dummy_coded = NA,
  group,
  keep_original = TRUE
)

Arguments

data	A data.frame or a data.frame extension (e.g. a tibble).
cols	Dummy-coded or effect-coded columns for group-mean centering. Support 'dplyr::dplyr_tidy_select' options.
dummy_coded	Dummy-coded variables (cannot be effect-coded) for L2 aggregated means. Support 'dplyr::dplyr_tidy_select' options.
group	the grouping variable. Must be character
keep_original	default is 'FALSE'. Set to 'TRUE' to keep original columns

Value

An object of the same type as .data. The output has the following properties: 1. Columns from .data will be preserved 2. Columns with L1 scores that are group-mean centered 3. Columns with L2 aggregated means (i.e., percentage) that are z-scored

Examples

z_scored_mlm_categorical(mlbook_data,cols='female_eff',dummy_coded='female_dum','schoolnr')

---