Contains the transformation rules from the IQB studies

This functions allows the transformation rules (anchor parameters, population reference values, cut scores) for the large-scale studies conducted at the IQB to be retrieved from a centrally stored database. The function arguments are used to specify in detail for which study, subject, competency areas, and mode the values are to be requested.

Usage

getTrafo(dataBase = "I:/Methoden/10_sonstige Materialien/trafo.rda",
         mode=c("paper","pc"), grade=c("primary", "secondary"),
         subject = c("math", "deu", "eng", "frz", "bio", "che", "phy"),
         domain = c("all", "GL", "ZO", "RF", "MS", "GM", "DHW", "ZA", "ME",
                    "FZ", "DZ", "lesen", "hoeren", "ortho", "sg",
                    "CE", "CF", "PE", "PF", "BE", "BF"),
         study = c("bt", "vera"))

Arguments

dataBase: Either the path to the R object containing the database, or the database object itself, which has already been loaded.
mode: Specify the desired mode, e.g., paper-based assessment or computer-based assessment. Note that the data for computer-based studies will be added to the database gradually as they become available.
grade: Specify for which grade level parameters are to be requested.
subject: The desired subject. Only for the science subjects (i.e., bio, phy, che), it is possible to use more than obe subject. See the examples for more details.
domain: The desired domain. One or several domains can be chosen. Use "all", if you want to list all domains for the selected subject.
study: Since some large-scale studies use different transformation rules than VERA, the specific study to which the transformation rules refer must be specified here. For some subjects (such as the science subjects biology, chemistry, and physics), however, there are no transformation rules for VERA. For other areas (such as the global domain in mathematics for the primary level), the transformation rules for Bildungstrend and VERA are identical.

Details

The function returns anchor parameters, the mean, and the standard deviation of the reference population, as well as cut scores for various subjects, domains, grade levels, and modes, in the format required by equat1pl or transformToBista. The values returned by the function can be used directly as arguments for the equat1pl and transformToBista functions. For example, the returned list object anchor can be used as an argument for the equat1pl function. The returned list objects refPop and cuts can both be used as arguments for the transformToBista function.

The information stored in the database is based on this file: I:/Methoden/10_sonstige Materialien/Transformationsvorschriften.xlsx

The following is a brief description of the domains stored in the database.

Mode: paper:

Deutsch primary: Originally, the scores from the 2007 norm study and re-standardization were used for this purpose. The re-standardization became necessary because the mean scores differed implausibly widely between the pilot study and the norm study. In the 2011 Laendervergleich, it became apparent that the link to the norm parameters was functioning poorly, particularly for the listening domain, so the Laendervergleich was used to establish a new norm reference. Orthografy was included in the Laendervergleich only in a self-weighted subsample excluding students with special educational needs (SEN). Since this student group is also part of the population but was not included in the sampling until 2016, the population mean from 2011 was shifted by the difference that would have resulted had SEN students been included in the sampling. As a result, the cut scores also had to be shifted. This procedure, known at the IQB as the "backward trend," results in non-integer cut scores. Due to psychometric problems and severe violation of local stochastic independence assumption, the "Sprachgebrauch" domain was not included in the Laendervergleich and Bildungstrend (BT) studies. Consequently, only the norm values used for VERA are available for this domain.
Math primary: There was no "backward trend" in mathematics. The cut scores for all domains are also identical here. The same applies to the mean and standard deviation of the reference population.
Deutsch secondary: The reference population for VERA is the 2008 norming sample. For the Bildungstrend (BT), the reference is the BT2015, not—as one might think—the Laendervergleich 2009. The reason, once again, is that no SEN students were included in the 2009 laendervergleich. This also leads to a "backward trend" and shifted cut scores, but only for the Bildungstrend, not for VERA.

Value

A list with four objects.

anchor: The data.frame with anchor parameters
refPop: A data.frame with mean and standard deviation for the reference population
cuts: A list with cut scores
info: Character string with short information about refrence population

References

Examples

# transformation rules for all domains subject deutsch, primary level
primDeu <- getTrafo(mode="paper", grade="primary", subject = "deu",
           domain = "all", study = "vera")
#> Warning: cannot open compressed file 'I:/Methoden/10_sonstige Materialien/trafo.rda', probable reason 'No such file or directory'
#> Error in readChar(con, 5L, useBytes = TRUE): cannot open the connection

# transformation rules for all science domains
secScien<- getTrafo(mode="paper", grade="secondary", subject = c("bio", "che", "phy"),
           domain = "all", study = "bt")
#> Warning: cannot open compressed file 'I:/Methoden/10_sonstige Materialien/trafo.rda', probable reason 'No such file or directory'
#> Error in readChar(con, 5L, useBytes = TRUE): cannot open the connection