Creates a regresion table in APA style with bootstrap confidence intervals

apa.reg.boot.table(..., filename = NA, table.number = NA,
  number.samples = 1000)

Arguments

...

Regression (i.e., lm) result objects. Typically, one for each block in the regression.

filename

(optional) Output filename document filename (must end in .rtf or .doc only)

table.number

Integer to use in table number output line

number.samples

Number of samples to create for bootstrap CIs

Value

APA table object

References

Algina, J. Keselman, H.J. & Penfield, R.J. (2008). Note on a confidence interval for the squared semipartial correlation coefficient. Educational and Psychological Measurement, 68, 734-741.

Examples

#Note: number.samples = 50 below. # However, please use a value of 1000 or higher # View top few rows of goggles data set # from Discovering Statistics Using R set.seed(1) head(album)
#> adverts sales airplay attract #> 1 10.256 330 43 10 #> 2 985.685 120 28 7 #> 3 1445.563 360 35 7 #> 4 1188.193 270 33 7 #> 5 574.513 220 44 5 #> 6 568.954 170 19 5
# Single block example blk1 <- lm(sales ~ adverts + airplay, data=album)
apa.reg.boot.table(blk1)
#> #> #> apa.reg.boot.table is a beta version. #> Block 1: Generating 1000 bootstrap samples #> Bootstrap for Delta RSQ in progress
#> #> #> Regression results using sales as the criterion #> #> #> Predictor b b_95%_CI beta beta_95%_CI sr2 sr2_95%_CI r #> (Intercept) 41.12** [21.32, 58.54] #> adverts 0.09** [0.07, 0.10] 0.52 [0.43, 0.61] .27 [.19, .37] .58** #> airplay 3.59** [2.95, 4.28] 0.55 [0.46, 0.63] .29 [.21, .39] .60** #> #> #> #> Fit #> #> #> #> R2 = .629** #> 95% CI[.56,.70] #> #> #> Note. A significant b-weight indicates the beta-weight and semi-partial correlation are also significant. #> b represents unstandardized regression weights. beta indicates the standardized regression weights. #> sr2 represents the semi-partial correlation squared. r represents the zero-order correlation. #> Square brackets are used to enclose the lower and upper limits of a confidence interval. #> * indicates p < .05. ** indicates p < .01. #> #>
apa.reg.boot.table(blk1,filename="exRegTable.doc")
#> #> #> apa.reg.boot.table is a beta version. #> Block 1: Generating 1000 bootstrap samples #> Bootstrap for Delta RSQ in progress
#> #> #> Regression results using sales as the criterion #> #> #> Predictor b b_95%_CI beta beta_95%_CI sr2 sr2_95%_CI r #> (Intercept) 41.12** [22.68, 56.86] #> adverts 0.09** [0.07, 0.10] 0.52 [0.43, 0.60] .27 [.18, .36] .58** #> airplay 3.59** [2.99, 4.22] 0.55 [0.46, 0.63] .29 [.21, .40] .60** #> #> #> #> Fit #> #> #> #> R2 = .629** #> 95% CI[.56,.70] #> #> #> Note. A significant b-weight indicates the beta-weight and semi-partial correlation are also significant. #> b represents unstandardized regression weights. beta indicates the standardized regression weights. #> sr2 represents the semi-partial correlation squared. r represents the zero-order correlation. #> Square brackets are used to enclose the lower and upper limits of a confidence interval. #> * indicates p < .05. ** indicates p < .01. #> #>
# Two block example, more than two blocks can be used blk1 <- lm(sales ~ adverts, data=album) blk2 <- lm(sales ~ adverts + airplay + attract, data=album)
apa.reg.boot.table(blk1,blk2,filename="exRegBlocksTable.doc")
#> #> #> apa.reg.boot.table is a beta version. #> Block 1: Generating 1000 bootstrap samples #> Block 2: Generating 1000 bootstrap samples #> Bootstrap for Delta RSQ in progress
#> #> #> Regression results using sales as the criterion #> #> #> Predictor b b_95%_CI beta beta_95%_CI sr2 sr2_95%_CI r #> (Intercept) 134.14** [118.96, 149.22] #> adverts 0.10** [0.08, 0.11] 0.58 [0.48, 0.66] .33 [.23, .44] .58** #> #> #> #> (Intercept) -26.61 [-54.63, 10.71] #> adverts 0.08** [0.07, 0.10] 0.51 [0.42, 0.59] .26 [.17, .35] .58** #> airplay 3.37** [2.72, 3.96] 0.51 [0.43, 0.60] .25 [.17, .34] .60** #> attract 11.09** [6.04, 14.82] 0.19 [0.10, 0.27] .04 [.01, .07] .33** #> #> #> #> Fit Difference #> #> #> R2 = .335** #> 95% CI[.23,.44] #> #> #> #> #> #> R2 = .665** Delta R2 = .330** #> 95% CI[.60,.73] 95% CI[.23, .43] #> #> #> Note. A significant b-weight indicates the beta-weight and semi-partial correlation are also significant. #> b represents unstandardized regression weights. beta indicates the standardized regression weights. #> sr2 represents the semi-partial correlation squared. r represents the zero-order correlation. #> Square brackets are used to enclose the lower and upper limits of a confidence interval. #> * indicates p < .05. ** indicates p < .01. #> #>
# Interaction product-term test with blocks blk1 <- lm(sales ~ adverts + airplay, data=album) blk2 <- lm(sales ~ adverts + airplay + I(adverts * airplay), data=album)
apa.reg.boot.table(blk1,blk2,filename="exInteraction1.doc")
#> #> #> apa.reg.boot.table is a beta version. #> Block 1: Generating 1000 bootstrap samples #> Block 2: Generating 1000 bootstrap samples #> sr2 CI lower bound(s) adjusted to zero as per Algina, Keselman, & Penfield (2008) #> #> Bootstrap for Delta RSQ in progress #> Delta R2 CI lower bound adjusted to zero as per Algina, Keselman, & Penfield (2008) #>
#> #> #> Regression results using sales as the criterion #> #> #> Predictor b b_95%_CI beta beta_95%_CI sr2 sr2_95%_CI #> (Intercept) 41.12** [21.93, 59.02] #> adverts 0.09** [0.07, 0.10] 0.52 [0.43, 0.60] .27 [.18, .36] #> airplay 3.59** [2.97, 4.21] 0.55 [0.45, 0.63] .29 [.20, .39] #> #> #> #> (Intercept) 28.30* [-3.47, 54.37] #> adverts 0.11** [0.06, 0.16] 0.69 [0.38, 0.97] .05 [.01, .09] #> airplay 4.02** [3.06, 5.17] 0.61 [0.47, 0.77] .15 [.08, .23] #> I(adverts * airplay) -0.00 [-0.00, 0.00] -0.19 [-0.52, 0.16] .00 [.00, .02] #> #> #> #> r Fit Difference #> #> .58** #> .60** #> R2 = .629** #> 95% CI[.56,.70] #> #> #> .58** #> .60** #> #> R2 = .632** Delta R2 = .003 #> 95% CI[.56,.70] 95% CI[.00, .02] #> #> #> Note. A significant b-weight indicates the beta-weight and semi-partial correlation are also significant. #> b represents unstandardized regression weights. beta indicates the standardized regression weights. #> sr2 represents the semi-partial correlation squared. r represents the zero-order correlation. #> Square brackets are used to enclose the lower and upper limits of a confidence interval. #> * indicates p < .05. ** indicates p < .01. #> #>