(07-11-2018, 07:23 PM)fredtoast Wrote: Perhaps you should tell all of the experts in the filed who reviewed this study that you know more about it s validity without even looking at it than they do.

They could not multiple people into interview for the same job claiming to have the exact same birthdate, hometown, and class ranking in the same college class.  But that is why these studies are posted for peer review.  If any of the variables would change the results then the study would be panned.  But you have never read it so you have no clue.

(07-11-2018, 07:21 PM)BmorePat87 Wrote: So I think all of the things you asked to be the same were.

(07-11-2018, 07:29 PM)Beaker Wrote: Except for race AND criminal record. 2 independent variables at least.

(07-11-2018, 07:28 PM)bfine32 Wrote: Who the hell said it wasn't a valid study? 

(07-11-2018, 07:32 PM)BmorePat87 Wrote: Comparing a criminal white to a non criminal black tester, there are two IV. Not that I think two makes us throw this out, but ok let's ignore that match up.

So let's just compare the non criminal white to non criminal black and the criminal white to criminal black. It met all of your requirements (same age, same looks, same personality, trained the respond the same way, exact same resume). Even those numbers are significant. 

(07-11-2018, 07:33 PM)Beaker Wrote: I did. If you are trying to call it scientific.

But other than that, the study still illuminates some problems. Which means it was useful to some degree.

(07-11-2018, 07:35 PM)Beaker Wrote: I agree the study has use...I am just saying it does not follow correct scientific methodology and therefore cannot be referred to as any kind of scientific conclusion.

Quote: Not to mention that it was only in Wisconsin, so to assume the rest of the nation behaves exactly the same way is also an invalid correlation.

(07-11-2018, 07:38 PM)BmorePat87 Wrote: What deviations from correct scientific methodology did you find when you read it?

(07-11-2018, 07:42 PM)Beaker Wrote: More than one independent variable. That alone invalidates it as a scientific study. (Via the scientific method)

(07-11-2018, 07:42 PM)Beaker Wrote: More than one independent variable. That alone invalidates it as a scientific study. (Via the scientific method)

Let me sum up my point:

The study has use and illuminates some problems for further discussion/study.

The study cannot be considered scientific due to incorrect methodology.

(07-11-2018, 07:46 PM)fredtoast Wrote: No it doesn't

https://answers.yahoo.com/question/index?qid=20070823110409AAaQk6o

(07-11-2018, 08:00 PM)BmorePat87 Wrote: Also, I haven't been able to find anything that says more than one independent variable means something isn't scientific.

(07-11-2018, 08:05 PM)bfine32 Wrote: For real. I've found dozens are sites that states that more that 1 IV dismisses the validity of a scientific study. just in the past couple minutes.  

(07-11-2018, 08:11 PM)BmorePat87 Wrote: I found something from NC State that is funded by the National Science Foundation stating that you can have multiple independent variables. I posted the link. 

(07-11-2018, 08:14 PM)bfine32 Wrote: Yeah, that study stated that you must find the relationship for EACH IV if you have more than one. So thanks for that. 

Quote:You still haven't found anything that states multiple IVs negate the findings of a scientific study 

(07-11-2018, 08:18 PM)BmorePat87 Wrote: Well the link wasn't a study, but the study that Fred referenced does find the relationship for each IV and then the relationships between the two.

You're welcome :)





You're right. I still haven't found anything that states multiple independent variables negates the findings of a scientific study.

But you're the one claiming that, not me....

Quote:The scientific method defines a set of practices and conventions which will tend to create increasingly accurate theories about how the world works. Experiments carried out according to the scientific method seek the effect one variable has on another. Isolating the dependent variable is important because it clarifies the effects of the process on the independent variable under investigation.

Quote:TL;DR (Too Long; Didn't Read)
Testing only one variable at a time lets you analyze the results of your experiment to see how much a single change affected the result. If you're testing two variables at a time, you won't be able to tell which variable was responsible for the result.