pand: Percentage of all non-overlapping data
In scan: Single-Case Data Analyses for Single and Multiple Baseline Designs

Description Usage Arguments Details Value Author(s) References See Also Examples

The pand function calculates the percentage of all non-overlapping data (PAND; Parker, Hagan-Burke, & Vannest, 2007), an index to quantify a level increase (or decrease) in performance after the onset of an intervention.

pand(
  data,
  dvar,
  pvar,
  decreasing = FALSE,
  correction = TRUE,
  phases = c("A", "B")
)

`data`	A single-case data frame. See `scdf` to learn about this format.
`dvar`	Character string with the name of the dependent variable. Defaults to the attributes in the scdf file.
`pvar`	Character string with the name of the phase variable. Defaults to the attributes in the scdf file.
`decreasing`	If you expect data to be lower in the B phase, set `decreasing = TRUE`. Default is `decreasing = FALSE`.
`correction`	The default `correction = TRUE` makes `pand` use a frequency matrix, which is corrected for ties. A tie is counted as the half of a measurement in both phases. Set `correction = FALSE` to use the uncorrected matrix, which is not recommended.
`phases`	A vector of two characters or numbers indicating the two phases that should be compared. E.g., `phases = c("A","C")` or `phases = c(2,4)` for comparing the second to the fourth phase. Phases could be combined by providing a list with two elements. E.g., `phases = list(A = c(1,3), B = c(2,4))` will compare phases 1 and 3 (as A) against 2 and 4 (as B). Default is `phases = c("A","B")`.

The PAND indicates nonoverlap between phase A and B data (like PND), but uses all data and is therefore not based on one single (probably unrepresentative) datapoint. Furthermore, PAND allows the comparison of real and expected associations (Chi-square test) and estimation of the effect size Phi, which equals Pearsons r for dichotomous data. Thus, phi-Square is the amount of explained variance. The original procedure for computing the PAND (Parker, Hagan-Burke, & Vannest, 2007) does not account for ambivalent datapoints (ties). The newer NAP overcomes this problem and has better precision-power (Parker, Vannest, & Davis, 2014).

`PAND`	Percentage of all non-overlapping data.
`phi`	Effect size Phi based on expected and observed values.
`POD`	Percentage of overlapping data points.
`OD`	Number of overlapping data points.
`n`	Number of data points.
`N`	Number of cases.
`nA`	Number of data points in phase A.
`nB`	Number of data points in phase B.
`pA`	Percentage of data points in phase A.
`pB`	Percentage of data points in phase B.
`matrix`	2x2 frequency matrix of phase A and B comparisons.
`matrix.counts`	2x2 counts matrix of phase A and B comparisons.
`correlation`	A list of the `correlation` values: statistic, parameter, p.value, estimate, null.value, alternative, method, data.name, correction.
`correction`	Logical argument from function call (see `Arguments` above).

Juergen Wilbert

Parker, R. I., Hagan-Burke, S., & Vannest, K. (2007). Percentage of All Non-Overlapping Data (PAND): An Alternative to PND. The Journal of Special Education, 40, 194-204.

Parker, R. I., & Vannest, K. (2009). An Improved Effect Size for Single-Case Research: Nonoverlap of All Pairs. Behavior Therapy, 40, 357-367.

Other overlap functions: corrected_tauSC(), nap(), overlapSC(), pem(), pet(), pnd(), tauUSC()

## Calculate the PAND for a MMBD over three cases
gunnar <- scdf(c(2,3,1,5,3,4,2,6,4,7), B.start = 5)
birgit <- scdf(c(3,3,2,4,7,4,2,1,4,7), B.start = 4)
bodo   <- scdf(c(2,3,4,5,3,4,7,6,8,7), B.start = 6)
mbd <- c(gunnar, birgit, bodo)
pand(mbd)
pand(bodo)

## Calculate the PAND with an expected decrease of phase B scores
cubs <- scdf(c(20,22,24,17,21,13,10,9,20,9,18), B.start = 5)
pand(cubs, decreasing = TRUE)