Subjectivist Approaches to Evaluation - Evaluation Methods in Biomedical Informatics - page 257

FIGURE 9.1. Natural history of a subjectivist study.
interested parties. By analogy to a contract, these ground rules can be
changed during a study with the consent of all parties. (Although essential
to a subjectivist study, a memo of understanding or evaluation contract is
recommended for all studies, irrespective of methods employed.)
2a. Immersion into the environment: At this stage the investigators begin
spending time in the work environment. The activities range from formal
introductions to informal conversations and the silent presence of the inves-
tigators at meetings and other events. Investigators use the generic term
field to refer to the setting, which may be multiple physical locations, where
the work under study is carried out. Trust and openness between the inves-
tigators and those in the field are essential elements of subjectivist studies.
If a subjectivist study is in fact to generate insights with minimal alteration
of the environment under study, those who live and work in the field (clin-
icians, patients, researchers, students, and others) must feel sufficiently com-
fortable with the presence of the investigators to go about their work in the
customary way. Time invested by the investigators in building such rela-
tionships pays compound interest in the future.
2b. Initial data collection to focus the questions: Even as immersion is
taking place, the investigator is already collecting data to sharpen the initial
questions or issues guiding the study. The early discussions with those in the
field and other activities primarily targeted toward immersion inevitably
begin to shape the investigators' views. Immersion and initial data collec-
tion are labeled “2a” and “2b” to convey their close interaction. Almost
from the outset, the investigator is typically addressing several aspects of
the study simultaneously.
3. Iterative loop: At this point, the procedural structure of the study
becomes akin to an iterative loop as the investigator engages in cycles of
data collection, analysis and reflection, and reorganization. Data collection
involves interview, observation, document analysis, and other methods.
Data are collected on planned occasions as well as serendipitously or spon-
taneously. The data are carefully recorded and interpreted in the context of
what is already known. Reflection entails the contemplation of the new find-