predictRaw {cNORM} | R Documentation |
Most elementary function to predict raw score based on Location (L, T score), Age (grouping variable) and the coefficients from a regression model. WARNING! This function, and all functions depending on it, only works with regression functions including L, A and interactions. Manually adding predictors to bestModel via the predictors parameter is currently incompatible. In that case, and if you are primarily interested on fitting a complete data set, rather user the predict function of the stats:lm package on the ideal model solution. You than have to provide a prepared data frame with the according input variables.
predictRaw( norm, age, coefficients, minRaw = -Inf, maxRaw = Inf, covariate = NULL )
norm |
The norm score, e. g. a specific T score or a vector of scores |
age |
The age value or a vector of scores |
coefficients |
The coefficients from the regression model or a cnorm model |
minRaw |
Minimum score for the results; can be used for clipping unrealistic outcomes, usually set to the lower bound of the range of values of the test (default: 0) |
maxRaw |
Maximum score for the results; can be used for clipping unrealistic outcomes usually set to the upper bound of the range of values of the test |
covariate |
In case, a covariate has been used, please specify the degree of the covariate / the specific value here. |
the predicted raw score or a data.frame of scores in case, lists of norm scores or age is used
Other predict:
derivationTable()
,
getNormCurve()
,
normTable()
,
predictNorm()
,
rawTable()
# Prediction of single scores normData <- prepareData(elfe) m <- bestModel(data = normData) predictRaw(35, 3.5, m$coefficients) # using a cnorm object result <- cnorm(raw = elfe$raw, group = elfe$group) predictRaw(35, 3.5, result) # Fitting complete data sets fitted.values <- predict(m) # break up contribution of each predictor variable fitted.partial <- predict(m, type = "terms")