get_results: Get results from an Ibex controller

In antonmalko/ibextor: Easy preprocessing of Ibex data

Description

The functions are all called get_results_X, where X corresponds to an abbreviation of an Ibex controller name:

• get_results_aj - AcceptabilityJudgment

• get_results_daj - DashedAcceptabilityJudgment

• get_results_ds - DashedSentence

• get_results_q - Question

• get_results_fs - FlashSentence

Usage

get_results_aj(file_name, elem_number = NULL, del_col = NULL,
  del_mode = "auto", col_names = NULL, partial_names = TRUE,
  col_classes = NULL, partial_classes = TRUE, short_subj_ids = TRUE, ...)

get_results_daj(file_name, elem_number = NULL, del_col = NULL,
  del_mode = "auto", col_names = NULL, partial_names = TRUE,
  col_classes = NULL, partial_classes = TRUE, short_subj_ids = TRUE,
  sprt = TRUE, ...)

get_results_ds(file_name, elem_number = NULL, del_col = NULL,
  del_mode = "auto", col_names = NULL, partial_names = TRUE,
  col_classes = NULL, partial_classes = TRUE, short_subj_ids = TRUE,
  sprt = TRUE, ...)

get_results_fs(file_name, elem_number = NULL, del_col = NULL,
  del_mode = "auto", col_names = NULL, partial_names = TRUE,
  col_classes = NULL, partial_classes = TRUE, short_subj_ids = TRUE, ...)

get_results_q(file_name, elem_number = NULL, del_col = NULL,
  del_mode = "auto", col_names = NULL, partial_names = TRUE,
  col_classes = NULL, partial_classes = TRUE, short_subj_ids = TRUE, ...)

Arguments

file_name	character. Path to the file with Ibex results
elem_number	Element number in IBEX. Only useful when a) You specified multiple controllers within one item and b) Some of those controllers have the same type. In this case you can specify which exactly of those duplicate controllers you need.
del_mode	how to determine which columns have to be deleted. Can be one of: “auto” (decide based on data), “user” (only take into account user's requests), “mixed” (take into account user request and data). 'auto' is used by default if del_col is not specified; if it is, “mixed” is used. See Details
col_names	character vector with names for the columns
partial_names	logical. If TRUE, the first seven columns will receive default names (“timestamp”, “md5_hash”, “controller”, “presentation_order”, “element_number”, “type”, “item”), since these columns are the same in all (default) Ibex controllers. So col_names will be taken to specify names for columns starting from 8. If FALSE, col_names will be taken as specifying names for all columns
col_classes	character vector with classes to be assumed for the columns
partial_classes	logical. If TRUE, the first seven columns will receive default classes: timestamp, md5_hash, controller, type, group - character, item_number, element_number - numeric, So col_classes will be taken to specify classes for columns starting from 8. If FALSE, col_classes will be taken as specifying classes for all columns
short_subj_ids	logical. If TRUE, each subject will be assigned a short index. If FALSE, a combination of timestamp and browser md5 hash will be used to identify each subject.
...	further parameters passed to read.csv
sprt	logical. Only applicable to “DashedSentence” and “DashedAcceptabilityJudgment” controllers. Set to TRUE if the mode was set “self-paced reading” in Ibex.Set to FALSE if the mode was “speeded acceptability”.

Value

data.frame with the preprocessed data associated with the requested controller

Column deletion

The first seven columns in IBEX results always are:

1. Time the results are received

2. MD5 hash of participant's IP

3. Controller type

4. Item presentation order

5. Element number

6. Type

7. Item

Notice that the names given here for column 4 (“Item presentation order”) and 7 (“Item”) are not exactly the names listed in the Ibex manual (where they are instead named “item_number” and “Group”. However, the names listed here better correspond to the nature of the data contained in the corresponding column. See ch.15 (“Terminological clarifications”) in Ibex manual for clarifying remarks.

By default, the column deletion mode is “auto”. It means that the function will remove columns not containing useful information. This means:

• “md5 hash” column is always removed since it's not usually used in analyses

• “controller” column is removed if there is just one controller in the data - it is assumed, the user will know what controller that is

• “element number” column is removed if that column only contains zeros

• “item” column is removed if it only contains NULL values

The above applies to all controllers. In addition:

• for “DashedSentence” and “DashedAcceptabilityJudgment” in “self-paced reading” mode column 11 (“line_break”) is removed if it only contains FALSE values (i.e. if all sentences in the experiment fit on one line and there were no line breaks).

• For “Question”, “AcceptabilityJudgment” and “DashedAcceptabilityJudgment” in “speeded acceptability” mode column 15 (“is_correct”) is removed if it contains only NULL valuess (if this is the case, it means that correct answers were not specified in Ibex, so this column is of no use).

In “user” deletion mode the data will not be taken into account at all, and only the columns specified by user will be deleted.

In “mixed” deletion mode the columns to delete are the union of the sets of automatically deleted columns and user-specified columns.

Specific functions

“DashedSentence” and “DashedAcceptabilityJudgment” have two modes of data presentation: “self-paced reading” and “acceptability judgment”. If an experiment used those controllers in both mode, the corresponding functions will issue a watning and only select results consistent with sprt parameter.

antonmalko/ibextor documentation built on May 25, 2019, 2:26 p.m.