Table 2.

Summary of guidelines for best practice functional analysis (FA).

Recommendation	Reason	Effect	Evidence
Prior to FA implementation
Rule out biological/medical events	May negate assessment and treatment	Ensures results are not related to a biological process	Kennedy and Meyer (1996) O’Reilly (1995)
Limit access to reinforcers outside of the FA	May abolish responding	Increase responding Limit false negative	McGinnis et al. (2010) O’Reilly et al. (2009)
Limit the number of target behaviors	Responses that contact reinforcement may abolish motivation for other responses	Increase responding Limit false positive Limit false negative	Asmus et al. (2003) Jessel et al. (2014) Beavers and Iwata (2011)
Prescreen for automatic reinforcement (when suspected)	May reduce ambiguity and make FA more efficient	Determines if behavior is maintained by automatic reinforcement	Querim et al. (2013)
During FA implementation
Use a fixed sequence	Limits potential carryover effects by providing control condition immediately following test	Enhance discrimination	Hammond et al. (2013) Iwata, Dorsey, et al. (1982/1994)
Use different SDs in each FA condition	Limits potential carryover effects by signaling the current condition	Enhance discrimination	Conners et al. (2000)
Allow problem behavior to subside before beginning the next session	Limits potential carryover effects	Enhance discrimination	McGonigle et al. (1987)
Use low preferred toys or no toys in the attention condition	Toys may compete with attention	Increase responding Limit false negative	Roscoe, Carreau, MacDonald, and Pence (2008)
Use attention form typically used	Ensures attention is relevant	Increase responding Limit false negative	Kodak, Northup, and Kelley (2007) Piazza et al. (1999)
Conduct demand assessment to identify tasks to include in the demand condition	Ensures demands are aversive	Increase responding Limit false negative	Call, Pabico, and Lomas (2009) Roscoe, Rooker, Pence, and Longworth (2009)
Include tasks typically used	Ensures demands are aversive	Increase responding Limit false negative	Asmus et al. (1999) Iwata, Pace, Kalsher, Cowdery, and Cataldo (1990) McCord, Iwata, Galensky, Ellingson, and Thomson (2001) McComas, Hoch, Paone, and El-Roy (2000), Smith, Iwata, Goh, and Shore (1995)
Use descriptive assessment to determine items in tangible condition	Ensure that tangible item is similar to delivery outside of the FA	Increase responding Limit false negative or positive	Rooker et al. (2011)
Use people as therapists who usually provide the consequence	Ensures that appropriate SDs and reinforcers are present	Increase responding Limit false negative	English and Anderson (2004) McAdam, DiCesare, Murphy, and Marshall (2004) Thomason-Sassi, Iwata, and Fritz (2013)
Post-session
Graph all target behavior separately	Ensures responding is related to the relevant stimulus conditions	Differentiation may be easier to observe	Derby et al. (1994)
Designate appropriate control conditions	Makes analysis easier	Differentiation may be easier to observe	Fischer, Iwata, and Worsdell (1997) Kahng and Iwata (1998)
Graph responding occurring when the motivational operation is present	Ensures responding is related to the relevant motivation	Differentiation may be easier to observe	Roane, Lerman, Kelley, and Van Camp (1999)
Use statistical guidelines for visual analysis	Ensures consistency in FA outcomes	Differentiation may be easier to observe	Hagopian et al. (1997) Roane, Fisher, Kelley, and Mevers (2013)