CRAN Release 0.10.4

Version 0.10.4

New features

• New demo data set (self-generated fake data): selfother (can be used to try inclusion of control variables and to try cubic RSA)

Bug fixes/Glitches

• Fixed error in estimation of SRSQD model (by replacing a model constraint by an equivalent formula).

CRAN version 0.10.2

Moved tkrplot package to "Suggests" to make RSA compatible with Apple Silicon.

CRAN release v0.10.1

Bug fixes

• Fixed error in aictab function (had to react to a change in R 4.0.0 concerning stringsAsFactors)
• Fixed error in discrepancy output of the summary() function

CRAN release v0.10.0

New features

• Started to implement unit testing with covr.

• New option claxes.alpha in plot(). When plotting one of the cubic models "CL" or "RRCL" with the axes K1 and K2, this new option can be used to change the alpha level for which the regions of significance (i.e., the positions of the lines K1 and K2 that demarcate these regions) are computed.

• New option alphacorrection in caRange(). Enables Bonferroni-correction when testing the outcome predictions of all data points behind the line E2 for the cubic models "CA" and "RRCA".

• Control variables can now be included in the RSA model by use of the option control.variables. When control variables are included in the model...

  • ...you have the option to center the control variables before model estimation, by use of the option center.control.variables. This can improve interpretability of the intercept, which will then reflect the predicted outcome value at the point (X,Y)=(0,0) when all control variables take their respective \emph{average` values.
  • ...the summary will show not only the overall R^2 of the model (which includes variance that is explained by the control variables), but also the increment of R^2 as compared to the baseline model with intercept and control variables. This R^2 increment will typically be of interest because it refers to the amount of variance explained by the two predictors X and Y (plus their squared and interaction terms) in the RSA model.
  • ...the AIC table obtained with aictab will include two additional columns: the increment of R^2 as compared to the baseline model (R2.baseline) and the p-value for F-test of this increment (R2.baseline.p).
  • ...the response surface that is shown with the plot function will show the model-predicted outcome values when all control variables take their respective mean values.
  • ...the number of parameters K per model that is shown in the aictab table will include the number of control variables (in addition to all freely estimated paramters, the intercept, and the residual variance).
    `

• New options center="pooled" and scale="pooled" in RSA(), which allow centering/scaling the predictor variables on their pooled mean/SD. This option is typically preferred over variable-wise centering/scaling, because the "pooled" version preserves commensurability of the predictor scales. The possible options are: Default option ("none") applies no centering. "pooled" centers the predictor variables on their \emph{pooled sample mean. "variablewise" centers the predictor variables on \emph{their respective sample mean. You should think carefully before applying the "variablewise" option, as centering the predictor variables at different values (e.g., their respective means) can affect the commensurability of the predictor scales.

Bug fixes/Glitches

• Had to react to a change in lavaan concerning model comparisons. In case that the models were estimated with robust ML and one of the models had df=0 (i.e., the full second-/third-order polynomial model), lavaan::lavTestLRT would not allow to compare the (df=0)-model to a nested model because no scaled test statistic had been computed for the (df=0)-model in this case. In the released version, the internal label of the test statistic of the (df=0)-model is overwritten so that the chi-square difference test will be computed anyway. This is valid because the model with zero degrees of freedom has a chi-square test statistic of T=0. The output of the comparison will be the scaled test statistic of the nested model, which is the correct statistic for the scaled chi-square difference test in this case. Note that lavaan versions older than 0.6-3 (i.e., before 09/2018) have provided the standard chi-square instead of the scaled chi-square test statistic of the nested model in such a situation. Results that were obtained with lavaan version <= 0.6-2 can be reproduced by setting estimator="ML" in RSA().
• The print function for RSA objects now defaults to a sensible global model (second- or third-order full polynomial model) if no specific model is provided.
• Fixed the formula for the AICc in aictab and compare2. In the formula of the (first-order) AIC, the number of free parameters K included the intercept and residual variance, but the second-order correction term did not. These two parameters are now consistently counted in K. This means that results for the (second-order) AICc might slightly change due to the fix, whereas results for the (first-order) AIC are the same.
• In the summary, the R^2 value of the full model is now the R^2 of the model that was estimated with lavaan. The R^2 of the lm model which was shown here before can still be extracted from the RSA() output object by inspecting $LM.
• Got rid of notorious "Warning sqrt(b3*b5)"

CRAN version v0.9.13

New features

• Add some additional security checks for parameters of the plotRSA function.

Bug fixes/Glitches

• Fixed the demoRSA function.
• Had to react to a change in lavaan concerning missing values. If you want FIML estimation, you need lavaan version >= 0.6.3; an error is printed if the version is below that.
• Fixed a bug where the percentage of (in)congruent cases was not printed when missing values were present.

CRAN version 0.9.12

New features

• New parameter axesStyles in plotRSA: Define custom styles for LOC, LOIC, PA1, and PA2. Recognizes lty, lwd, and col. If you define a style for an axis, you have to provide all three parameters, otherwise a warning will be shown.
• The cubic models CA, RRCA, CL, and RRCL have been implemented (see Humberg, Nestler, Schönbrodt, & Back, in preparation, for details on how to use these models) and the plot() function has been adapted respectively. The new functions caRange() and clRange() have been implemented. They should be applied when testing cubic models to ensure that the model is interpreted for the whole range of realistic predictor combinations.

Bug fixes/Glitches

• The newly introduced a5 parameter was displayed as a second 'a1' in the 3d plot - this has been fixed.
• When counting the rate of discrepant predictor combinatios, standardize the predictor difference at the predictors' grand standard deviation. This ensures that the percentage of (in)congruent predictors is independent of common linear transformations of the predictor variables. Accordingly, a predictor combination is now categorized as incongruent when X and Y differ by more than half a grand standard deviation from each other.

CRAN release 0.9.11

New features

• New param in plotRSA: suppress.grid.
• If link="probit" in plotRSA, the 3d plot automatically scales the z-axis to c(0, 1)
• If link="probit" in plotRSA, the raw data points now can take the predicted probability (plotRSA(..., points=list(value="predicted")))
• cex params in plotRSA slightly changed: now we have three cex parameters to independently control the font size of the main title (cex.main), the axes labels (cex.axesLabel), and the tick labels (cex.tickLabel)
• New surface parameter for testing strict congruence patterns ("a5"). Thanks to Sarah Humberg for implementation.

Bug fixes

• Fixed a wrong formula in the standard error of a4 (the formula was corrected in an erratum of Shanock, L. R., Baran, B. E., Gentry, W. A., & Pattison, S. C. (2014). Erratum to: Polynomial regression with response surface analysis: A powerful approach for examining moderation and overcoming limitations of difference scores. Journal of Business and Psychology, 29, 161. http://doi.org/10.1007/s10869-013-9317-6. Thanks to Sarah Humberg for pointing to the erratum).
• Fixed a bug in the curvature of the principal axes (closing issue 2 on Github; thanks to scith for reporting the bug).

CRAN 0.9.10

• Fixed a wrong sign in the C parameter of the SRR model (thanks to Sarah Humberg for detecting the bug.)

CRAN release v0.9.9

• Better documentation for RSA.ST function.
• Made necessary changes due to lavaan update.
• Fixed issue due to ggplot2 update.

CRAN release 0.9.8

Version 0.9.8

Bug fixes

• Removed the beta version of the control.variables feature - this broke some functions.
• Removed some unnecessary warning that might confuse users.

Version 0.9.7

New features and improvements

• Improved aictab-plot; now also has a black/white option (bw=TRUE)
• Added evidence ratio to the aictab output: How much more likely is the best model (in first row) compared to each other model in the set?
  • compare, compare2, and aictab gained a digits argument
• Reduced the set of models that is computed by default in the RSA function. Uncommon models, such as the mean model or the diff model have been removed.
• Removed eigenvalues from standard summary. Most users will never need them.
• When CIs and p values are bootstrapped, S is automatically tested against 1 (which is the null hypothesis).
• Improved contour plot (bag plot is now better visible; if xlim and ylim are set, the range is not extended anymore to include all points).
• New option in plotRSA: showSP (TRUE/FALSE) controls the display of the stationary point in contour plots, showSP.CI (TRUE/FALSE) controls the display of the stationary point's confidence interval.
• New demo data set: motcon2

Bug fixes

• Fixed a bug that broke the surface="smooth" option in plotRSA
• Fixed a bug that out.rm=FALSE did not work any more.
• Fixed a bug where getPar(..., type="r2.p") returned a wrong p value (the output in print and model comparisons always was correct.)
• Fixed wrong label in aictab plot