An Improved Effect Size for Single-Case Research: Nonoverlap of All Pairs

A comparison of mean phase difference and generalized least squares for analyzing single-case data

“…Several modifications of the PND were proposed in order to avoid relying on a single baseline data point. The proposal that has proven (Manolov, Solanas, Sierra, & Evans, 2011;Parker & Hagan-Burke, 2007) to perform best of these alternatives to the PND is the Nonoverlap of all pairs (NAP; Parker & Vannest, 2009), as it is not affected by serial dependence and also distinguishes between presence and absence of effect. This procedure compares each baseline datum to each treatment datum to quantify the percentage of nonoverlap (i.e., the degree to which the treatment measurements are improved as compared to the baseline measurements).…”

Técnicas para el análisis de diseños de caso único en la práctica clínica: ejemplos de aplicación en el tratamiento de víctimas de atentados terroristas

“…Este intervalo de confianza permite evaluar el NAP obtenido con mayor cautela, siendo consciente de sus limitaciones en función del número de datos con el que ha sido obtenido, a la vez que permite estimar si dicho NAP difiere de forma estadísticamente significativa de .50 o 50%, lo cual ocurriría cuando el límite inferior de su intervalo de confianza al 95% fuera mayor que .50 e inferir, en consecuencia, si se ha producido en la fase de tratamiento un cambio de nivel respecto a la LB. El cálculo del intervalo de confianza también es posible para el PEM mediante la prueba binomial de T e l e f ó n i c a P r e S e s i ó n 1 S e s i ó n 3 S e s i ó n 5 S e s i ó n 7 S e s i ó n 9 S e s i ó n 1 1 S e s i ó n 1 3 S e s i ó n 1 5 P o s t T e l e f ó n i c a P r e comparación de una proporción con la proporción teórica de .50 que representaría la mediana (Parker y Vannest, 2009). Para interpretar los valores del NAP, Parker y Vannest (2009) han propuesto unos valores de referencia que pueden consultarse en la tabla 1.…”

“…PEM assumes that the median is a good summary for baseline data, but it is not the case for the data often seen in single case design. NAP was developed mainly to improve upon PEM (and PND) by individually comparing all baseline and treatment phase data points (Parker & Vannest, 2009).…”

Small N designs for rehabilitation research

“…Promising nonparametric methods also exist for analyzing single-case data, for example, logistic regression [11]; nonoverlap of all pairs (NAP) [43]; and Tau-U, which combines nonoverlap of data and trend [39]. The advantages of the logistic method are that it does not assume a linear relationship between the independent and dependent variables and it does not require normally distributed variables or equal variance per cell (for example an AB design may be seen as a 2 × 2 table).…”

“…It may be calculated by hand but is also easily obtained as the area under the curve (AUC) from a receiver operator curve (ROC) analysis. NAP (or AUC) may be defined as "the probability that a score drawn at random from a treatment phase will exceed (overlap) that of a score drawn at random from a baseline phase" or as "the percent of nonoverlapping data between baseline and treatment phases" (p. 359) [43]. NAP scores range from 0.5 to 1.0 but may be transformed to have a range from 0 to 1.0 to represent deterioration in behavior or treatment [44].…”