Interrogating NUll effects PDF

Preview

Summary

These are just copies of what the teacher provided us...

Description

Interrogating Null Effects: What if the independent variable does not make a difference Null effects Can happen in a within groups design or a pretest and post test design and even in a correlational study. This means that the independent variable manipulated by the experimenters did not result in a change in the dependent variable. Sometimes it might be the case at the independent variable really did not affect the dependent variable. it could mean that our theory is incorrect. Another possible reason for another fact is that the study was not designed or conducted carefully enough. There might have not been enough between groups difference, or there might have been too much within groups variability. Between groups difference. Examples of not having enough between groups difference would be weak manipulations, and insensitive measures, ceiling and floor effects, and reverse design confounds. This might prevent study results from revealing a true difference that exist between two or more experimental groups. Weak manipulations: - It is important to ask how the researchers operationalize the independent variable. In other words you have to ask about the construct validity. Insensitive measures - This means the researchers have not use an I’ll for rationalization of the dependent variable with enough sensitivity. - When it comes to dependent measures it’s smart to use ones that I have detail, quantitative increments, not just two or three levels. Ceiling and floor of facts and independent variables - Sometimes the various levels of the independent variable would appear to make no difference. Ceilings and floors and dependent variables - Same thing as independent variables as being poorly designed. The most important thing to consider when interrogating a study with a null effect, it is important to ask how the independent independent variable’s were operationalized. MANIPULATION CHECKS. Design confounds acting in reverse - A study might be designed in such a way that a design confound actually counteracts, or reverses, some true effect of the independent variable. Too much within groups variability - Noise. - Measurement error - Individual differences. - Situation noise - Solution: Power Maybe there is just no effect to find? Consider that.

Noise - The greater the overlap, the smaller the effect size, and the less likely to group means will be statistically significant; that is, the less likely the stuff he will detect covariance. - When the data show less variability with in the groups, the effect size will be larger, and it’s more likely the main difference will be statistically significant. The less within Group variability, the less likely it is to hobscure a true group difference. Measurement Error - Use reliable, precise tools. - Measure more instances. Individual differences - Change the design - Add more participants Situation noise - Control the surroundings of the experiment....