I agree that the tense of the conclusion is not the most important thing-it is certainly possible to have a correct conclusion in the past tense.
Here are some examples:
* If a presidential approval poll was done in August 2009 and we constructed a CI for the president's approval rating, it would be wrong to make a generalization about "the proportion of US adults who said they approved of the president's performance in August 2009" because this is clearly referring to the sample.
* A better, but somewhat ambiguous conclusion would be about the "proportion of US adults who approved of the president's performance in August 2009." In this case, it is unclear if the group that "approved" is the sample or the population. Saying "all US adults" would make this much better, even in past tense.
* A better conclusion would be about "the proportion of all US adults who would have said "approve" had they been asked about the president's job performance in August 2009."
My point in the original email was that when a conclusion/interpretation is in the past tense, this is often a clue that the student might be misunderstanding the difference between the statistic and the parameter. But, you are right that the most important thing is to clearly identify the population/parameter regardless of the tense.
From: Dennis Roberts [mailto:firstname.lastname@example.org] Sent: Wednesday, June 22, 2011 6:35 AM To: AP Statistics Subject: RE: [ap-stat] Things I learned from the 2011 AP Statstics reading
At 01:05 AM 6/22/2011, Josh Tabor wrote:
Hello All- Notice that the tense of the verb is no longer in the past tense. While I am sure there are exceptions to this rule, I suggest to my students that interpretations/conclusions should never be in past tense because this often means that the student is referring to the data from the study rather than what they are trying to make an inference about.
Not to toss hmmms into this discussion but ... since the CI is based on data from a sample that already exists ... then the conclusion would seem to be about the overall population that existed when these data were collected. This is not a future sample ... on which we can say things about a population in the future ... it's about a sample now ... or in the past .. and it's associated population.
In a very real sense, the CI is about a past tense.
I don't think I would be worrying about the tense of the verb rather, I would be wanting to make sure that the student refers to the population from which this sample was taken.