## NCA limitations

Can NCA predict the presence of the outcome?

No. NCA only can predict the absence of an outcome, not the presence of the outcome. NCA focuses on single conditions (and their combinations) that prevent the outcome to occur. Traditional sufficiency methods such as Multiple Regression, Structural Equation Modeling, Partial Least Squares, as well as methods like QCA consider the complete causal structure that produces the outcome. These methods must be used to predict the presence of the outcome from a set of conditions.

Can NCA solve the problem of measurement error?

No. Just like other data-analysis approaches NCA presumes that the data to be analysed are valid and reliable (and calibrated). If this assumption is not correct the results of the NCA analysis can be flawed. When NCA is used for describing the sample or for making point estimates of the population from which the sample is drawn, NCA is not sensitive for measurement error of observations (far) below the ceiling line, but sensitive for measurement error of observations around the ceiling line.

Can NCA handle outliers?

No/Yes. If the outlier is caused by measurement error, the results of an NCA may be flawed (see measurement error). If the outlier is a real phenomenon, NCA can handle this outlier: if the outlier is below the ceiling, NCA results are not affected; if the outlier is above the ceiling, NCA takes the outlier into account (the ceiling includes the outlier). Note that all cases around the ceiling are “best cases”: with relatively low input (X) they achieve a relatively high output (Y).

Can NCA prove causality?

No. Just like other data analysis techniques, NCA alone cannot prove causality. It depends largely on the research design and the available theory whether or not it is plausible that the condition is a necessary cause. Experimental designs are preferred for that purpose. Observational studies need solid theory to infer causality.

Can NCA solve the problem of convenience sampling for statistical inference?

No. Just like other data-analysis approaches NCA presumes that the sample is a probability sample (e.g., random sample) from the population. Although this requirement is seldom met, if this assumption is not true the results of the NCA analysis (and any other data analysis approach for statistical inference) can be flawed.