INSTRUCTOR: Welcome to Enrichment Presentation Activity 3. In this lesson, we take a look at something called the hypothesis.

By definition, a research hypothesis is a specific and verifiable prediction regarding the relationship between or among two or more variables. Another way to look at it, it is a proposal that is to be tested and evaluated.

The hypothesis statement asserts that a relationship exists, and it also states the direction of that relationship. In other words, you as a researcher believe that one or more things are causing something else. For example, giving people a raise makes them happy. Or playing music while taking a test causes distraction. And so you make this into a prediction that you're going to explore with your research.

Where does the hypothesis come from? It can be an area of interest or a topic that you want to know more about. It can be an observation that you have made and want to investigate further. You start broadly and make it specific to a situation. Is this about all people or does this only happen in certain situations or under certain conditions?

We don't prove or disprove our hypothesis. Rather, we seek to find evidence to support our claim or proposal. When enough research is done that finds support, we accept our hypothesis as basically being true.

This is where it gets a little confusing. We start by pretending that our prediction is not true. So we might say that giving people a raise has no effect on happiness. When we do this we call this no-change statement the null hypothesis. The null means that there will be no change between the mean or average measurement of the first group pre-change and the mean of the second group post-change. Remember we are studying a group of people, so we look at the average measurement.

We then write up the opposite. There is a change, and we call it the alternative hypothesis. We are predicting that there will be a change in the means due to a manipulation we propose. Our manipulation in this example is that we are increasing salaries. If there is a measurable change as a result of our study, we will reject the null statement, which said that there was no change, and accept the alternative hypothesis. And we have a research result to share with whomever may be interested.

Here's an example of hypothesis statements. A null hypothesis statement might be that there will be no difference or change in productivity between workers who work day shift and those who work night shift. One alternative hypothesis may state there will be a decrease in productivity in night shift workers as compared to day shift workers. Maybe because we believe they goof off more at night.

Or another prediction might be that night shift workers are more productive than day shift workers, and that might be because there aren't as many distractions at night. Again, if you find a measurable change once you conduct your experiment-- in this case, monitoring production-- then you reject the null statement. And that means that you will accept your hypothesis statement. Remember, the hypothesis is your prediction about the effect of your manipulation.

For your assignment this week, write an observation about something. It can be something work- or school-related, or it could be something very simple, such as, the sky seems very blue today and it rained last night. Then, make a prediction out of your observation. For example, I might say that I believe the sky is bluer after a night of rain. Then write this in the form of null and alternative hypothesis statements.

Finally, what does it mean to reject the null hypothesis? Explain that.

If you have any questions, please let us know.