Checklists vs. Rating Scales

Observations vs. interviews Observations, as the name suggests, involve the observation of the study subjects, in a natural or an experimental environment. The researcher takes notes upon what he or she sees, either in a prominently-displayed position or in a less obtrusive position. Interviews may be unstructured, semi-structured, or highly-structured, but all types involve directly soliciting information from subjects (Aiken & Groth-Marnat & 2006). There are potential benefits and drawbacks to both observations and interviews. In terms of validity, observer bias and lack of awareness can corrupt observational data.

However, interviews can likewise be problematic given that the questions can intentionally or unintentionally be used to shape the respondents' replies in a particular and inaccurate fashion. Observations can be very difficult to reproduce, given they take place at specific moments in time and both methods make use of relatively small test populations, versus quantitative studies. This narrows the demographic reach of these largely qualitative methods. The advantage of observation is that it allows for subjects to speak for themselves and minimizes researcher interference.

For a population about which the researcher knows very little, such as in the case of an anthropological study of a foreign culture, observations may be more appropriate. Also, some subjects because of cultural or linguistic barriers might not be suitable to interview. However, interviews can still yield important insights about phenomena, particularly thoughts and feelings that cannot be readily observed. Interview responses to questions can also be more easily coded and compared with previous research than observations.

Ethical issues may arise with both that must be addressed, however, such as the consent of the observed to allow themselves to be subjected to a research study (which is more easily obtained in an interview format from either the subjects or, less ideally and more problematically, the parents or caregivers of the subjects if minors or individuals unable to consent are involved). Checklists vs. rating scales "Checklists usually offer a yes/no format in relation to student demonstration of specific criteria" ("Assessment strategies and tools," 2014).

Either something is exhibited or not while rating scales allow observers "to indicate the degree or frequency of the behaviours, skills and strategies" ("Assessment strategies and tools," 2014). Rating scales allow for more ambiguity, which can be useful in terms of qualifying a response. However, the greater subjectivity allowed with rating scales can also mean they are potentially less reliable and consistent. It can be more difficult to assess a rating scale given the degree to which the number depends upon the subjective bias of the observer.

The validity of either method vs. another will likely depend upon the phenomenon being studied. For complex, multi-faceted conditions with many subtle shades of meaning, it is virtually essential to use some type of rating scale. In contrast, simpler concepts might benefit from a checklist. Checklists, in general, can be more easily compared with other studies, but again, the phenomenon must be very straightforward in nature for this to be accurate. For example, observing behaviors in a patient with OCD might be appropriate for a checklist vs.

assessing mood in a depressed patient which requires a rating.