Question: module 6 dq 1 argument response Rachel

Dolma (2010) defines the unit of analysis as "the entity that is being analyzed in a scientific research". Further, Dolma (2010) emphasizes that the unit of analysis should align with respect to what we want to generalize the results to. Since the researchers want to determine these behaviors and generalize these results by developing these behaviors in future physicians, the unit of analysis would be the behaviors of the physicians with high patient satisfaction ratings. However, the observational unit in this case would be different from the unit of analysis. Specifically, the observational unit is the encounter of the physician-patient dyad. Data could be collected not only from the video recordings but also through follow-up interviews with the individuals in the dyad as there may be behaviors that may not have been typical from the physician's perspective, or the physician's behavior could be predicated upon the patient's behavior. Furthermore, the various observed physician behaviors may be perceived by the patient differently than the researcher's interpretation of the observed behavior.

Estimating the number of data units can be difficult, because the number of physician behaviors observed in each encounter can vary by physician, patient, and the reason for the patient's visit. However, reflecting upon my personal experiences the types of behavior frequently observed in physicians include multiple forms of verbal and non-verbal communication between the physician and patient. If medical procedures or tests are performed during the visit, multiple other behaviors may be observed including types of physical touch (gentle versus rough), communicating before, during, and after the procedure to alleviate patient stress, and asking the patient for feedback throughout. Thus, if the number of observational units are a total of 50 dyads (5 patient-physician dyads for each of the 10 physicians) and the number of distinct behaviors observed are at least 5 per dyadic encounter, there could be at least 250 data units, in addition to the real potential that specific behaviors may occur more than once during each encounter, which could significantly increase the number of data units. Thus, the number of anticipated data units seems to be sufficient for this study, especially if other types of data are collected as mentioned above.