Theunits of analysis, in the realm in social science research, refers tothe “who” and “what” that forms the subject of the study.According to Bhattacherjee (2012), the most common units of analysisinclude individuals as well as groups, organizations, and socialartifacts. It is critical that differentiation is made between theunit for analysis and the unit of observation (Gale et al., 2013). Inthe present case, the behaviors of physicians, which led to highpatient satisfaction ratings, are the units of observation. However,the unit of analysis would be the physicians who were studied. Thenumber of recordings done is five while the physician-patientencounters are from 10 physicians. As such, the data units would be20 as each patient is likely to have encountered two physicians. Eachpatient having encountered two different physicians brings the numberof data units to twenty, as each of the encounters would form part ofthe data.
Theestimated number of data units seems adequate because the researcherwill be able to communicate satisfactory information, which wouldprovide sufficient meaning. Notably, Kahn et al. (2012) posits thatthe adequacy of the data units that will provide similar measurementswhen made repeatedly is the concern. In this case, studying twophysicians who attended to the same patient will most likely producesimilar results. However, a study of physicians who attended todifferent patients would not bring about similar results (Kahn etal., 2012). As such, it is critical that when conducting a study, anyelement that would most likely present similar observations should bereflected. This will ensure that similar themes in the data areestablished and a conclusion made as to what the similarity depicts(Kahn et al., 2012).
Bhattacherjee,A. (2012). Social science research: principles, methods, andpractices.
Gale,N. K., Heath, G., Cameron, E., Rashid, S., & Redwood, S. (2013).Using the framework method for the analysis of qualitative data inmulti-disciplinary health research. BMCmedical research methodology,13(1),117.
Kahn,M. G., Raebel, M. A., Glanz, J. M., Riedlinger, K., & Steiner, J.F. (2012). A pragmatic framework for single-site and multisite dataquality assessment in electronic health record-based clinicalresearch. Medicalcare,50.