What type of study is described? (NOT A MULTIPLE CHOICE)
a. Managers of a forest preserve want to know what percent of the visitors with dogs keep their dogs on a leash. The park assigns an employee to count the numbers of visitors that do and do not use a leash.
b. A local newspaper polls citizens of a city about whether they support a local tax levy?
Answers:
===========================================================
Explanation:
Part (a)
We have an observational study because the park ranger is observing the people coming and going, as to whether they leash their dog or not. The people being counted are not asked to fill out a form or survey, so that’s why we aren’t conducting a survey here.
Ideally, the park ranger should somehow make a visual note of each person so they do not count that person more than once. For example, let’s say we had 500 people go through the park. Of that 500, the true number of people who leash their dogs is 20. I don’t know what the true percentage is, but let’s just go with this hypothetical situation. If those 20 people pass back and forth where this manager is taking note, then he might overcount some individuals and inflate that 20 to say 40 or 60, or perhaps even 100. This is if the ranger doesn’t look at the individuals closely and simply marks a tally for “leash” vs “no leash”. So that’s why there needs to be some quick method to distinguish between any unique guest of the park.
——————————–
Part (b)
In contrast to part (a), we have a survey here this time. Each person is asked directly about a specific narrow well defined question. The newspaper reporter isn’t going out into the field to simply watch or listen to people discuss about taxes, so that’s why we don’t have an observational study going on. The reporter should try to keep the question as bias-free as possible. Also, they need to ensure that the sample is representative of the population.
In neither part (a) nor part (b) is an experiment being done. An experiment requires two groups: a treatment group and control group. Also, the researcher would directly affect the treatment group to somehow change one specific aspect, while keeping the control group the same (though possibly giving a placebo of some kind). For each of these, the researcher is simply trying to stay out of the picture as much as possible while gathering the external data.