Mk. Bengali et Kj. Ottenbacher, THE EFFECT OF AUTOCORRELATION ON THE RESULTS OF VISUALLY ANALYZING DATA FROM SINGLE-SUBJECT DESIGNS, The American journal of occupational therapy, 52(8), 1998, pp. 650-655
Objective. Single-subject research designs are used to conduct clinica
l research and outcome evaluation in occupational therapy. Confusion e
xists regarding the best method to analyze and interpret single-subjec
t data. Method. One hundred graphs displaying the results of published
single-subject research were examined to determine the influence of a
utocorrelation on the visual inferences made by the original investiga
tors. The graphs were selected fi om 20 articles published over 10 yea
rs in seven rehabilitation journals. The data were extrapolated and la
g 1 auto-correlation coefficients computed for both the baseline and t
reatment phases. Results. Data analysis focused on two issues: (a) whe
ther a relationship existed between the amount of autocorrelation pres
ent in a graph and the conclusion on the basis of visual analysis and
(b) whether the amount of autocorrelation varied across different phas
es of the single-subject graphs. When a significant degree of autocorr
elation was present, researchers using visual analysis were more likel
y to conclude that there war no clinically significant change in perfo
rmance. Autocorrelation values were significantly higher in the treatm
ent phases of the single-subject designs. Conclusion. Additional resea
rch is needed to establish a set of decision rules to assist clinician
s in using visual analysis to evaluate the results of single-subject r
esearch.