Video: Types of Statistical Studies

Is the following a survey, an observational study, or an experiment? A company has developed nontoxic oil paints. To investigate whether this new, environment-friendly product has attracted new clients, client and sale data were analyzed.

02:43

Video Transcript

Is the following a survey, an observational study, or an experiment? A company has developed non-toxic oil paints. To investigate whether this new, environment-friendly product has attracted new clients, client and sale data were analyzed.

So, in order to answer this question what we need to first look at is what a survey, an observational study, or an experiment are. Well, first of all, we’ll take a look at a survey. And a survey is data collection from a predefined group who give answers or insights on specific or varied topics.

So, if we take a look at an observational study. This is where subjects are observed and results are measured but independent variable is not controlled by investigator. And this is often for ethical or logistical reasons. For example, if we were looking into the effect of air quality on lung capacity, then it would not be ethically sound to give one group bad quality air while the other group had good quality air. Because, yes, that could show the difference between these types of quality air on lung capacity. However, giving one set of people within the study bad quality air would not be ethically sound. So, we could not use that kind of experiment.

And then finally, for an experiment, well, in an experiment, investigators control variables and observe outcomes. So, it has similarities to the observational study, but it’s in a much more controlled environment.

So, now what we need to do is work out whether our scenario is a survey, an observational study, or an experiment. So, let’s remind ourselves what the scenario is again. So, it’s a company who have developed a non-toxic oil paint. And they want to see whether the new product has attracted new clients. And to achieve that, what they’ve done is analyze client and sale data. Well, this seems more aligned to an observational study.

Because it’s not a survey, because they haven’t sent anything out and they aren’t asking specific questions of their clients. And it’s not an experiment, because they’re not controlling the variables. And this is because if we think about the scenario, then there have been a number of factors to whether the sales increased or decreased over a certain period. And they haven’t controlled all of these, so therefore, it wouldn’t be an experiment. So therefore, we can say that the scenario we’re looking at is an observational study.

Nagwa uses cookies to ensure you get the best experience on our website. Learn more about our Privacy Policy.