 Now we have our research question. We've done our literature search. We need to know what the variables would be required to do our research. Now the first thing would be to do a power calculation. I'm not going to go through that. That is for us to try and determine how many patients we would need for us to be able to have some power in our statistical analysis. That's a very important step and perhaps we'll make some extra videos just on that. We'll have to identify the patients from some sort of register. We might have our hospital register, the ward register where these patients are admitted, where diagnoses are written and that is where we're going to get our sample set from. And we have to now collect data on the type of infection. So we'll have to either classify it as major or minor and that is a nominal categorical data type. If I were just to write my data points under the variable type of infection as minor, major, minor, major, minor, minor, major. Those data points, the data type of those data points are nominal categorical. Categorical because minor is a word. I can't do mathematics using a word. It's nominal because there's no natural order to it. Now major is worse than minor but there really isn't a natural order sequence to these. The next data we're going to collect is gender. Now that's not part of our original research questions but you'll see on most clinical research papers there's a bit of the background information of the patients that we are dealing with in gender and age. Usually two of the common things that we do list. So once again we'll have two data points just to keep things simple here. We'll either classify sample patients as either male or female and once again you can well imagine this is nominal categorical data. Next age comes up. We'll just make note of each patient's age and we might use that in a little comparison as well although that was not part of our original research question. And that is a ratio type numerical continuous data type. Ratio type because there is a true zero. Someone can be zero years of age. It is numerical because we can do mathematical equations using those data points and it's a continuous data type. The admission HBA1C that's also going to be a ratio type numerical continuous data type and admission CRP also it's a value also with a true zero so that would also be a ratio type numerical data type which is also continuous. Important when we think about these variables the data points that we are going to collect for these variables it's important to think about what type of data that is because that will determine what kind of statistical analysis we can do on those data points when we get to Julia. Next up we'll set our hypotheses.