Convenience sample equivalent of "prevalence"

Question

EDIT: To clarify, my original (sleep-deprived) post:

We used the term 'prevalence' to describe the proportion of patients with a certain characteric in a cross-sectional convenience sample. A collaborator expressed concern regarding the use of the word 'prevalence' outside of the context of the 'population prevalence' or an estimate thereof (as might be obtained using a probability sampling method).

My question was - is it appropriate to use the term prevalence in relation to the sample prevalence (even in the case of a convenience sample). My feeling is strongly yes, but I would like some external validation and Google searches have not yielded exactly the answer I was searching for.

Answered below in the comments of the accepted answer.

=============================

What is the appropriate terminology to use when referring to the proportion of participants in a convenience sample that carry a given characteristic?

For example, I stood outside an ice cream parlor and asked everyone that walked in over the course of a day whether they preferred a cone or a cup. If 60/100 people responded "cup", could I reference this result as a 60% "prevalence" of individuals that preferred a cup for ice-cream while still acknowledging all of the caveats that come with a convenience sample (e.g. not generalizable, prone to sampling error, etc. etc.)

This is one of those things that the more I think about it, the crazier it makes me feel, and is relevant because a collaborator had objections to the use of the word "prevalence" or even "crude estimate of prevalence" (for a completely different study topic), meanwhile there are numerous examples of the word "prevalence" used to refer to the results of convenience samples published in our target journal.

Answer 1

For a proportion of individuals with some characteristic, there are two terms we can consider: prevalence and incidence. The distinction relies in how prevalent cases are counted. Prevalence includes both new and prevalent cases. Incidence only includes new (incident) cases. Therefore, you should consider the characteristic in the context of your work. Whether the characteristic is prevalence or incidence depends on the provided context. Often incidence requires multiple measures (since you need to identify who does not have the characteristic at baseline).

Example: let's consider the proportion of right-handedness. If we were collect handedness at birth (or as soon as it can be measured) that would correspond to incident handedness since all are new instances of right-handedness. However, if we collect handedness among adults then that would correspond to prevalence since both people who were right-handed at birth and became right-handed would be part of the proportion. However, you could also define right-handedness to be incident among adults (i.e. you would only look at people who had previously been left-handed and see if they began using their right-hand instead).

Returning to your ice cream example, I would consider it prevalence. The 60% constitutes both people who recently began to prefer 'cups' and people who have always preferred 'cups'. To find the incidence, you would take the 40% who responded as preferring 'cone' and conduct a follow-up survey.