0:04
So once data has been collected and analyzed,
you need to ask yourself how valid is this study?
Validity is just a fancy statistical word for truth.
It tells you how much you can trust the results of a particular study.
Over this course, we'll talk about a number of different types of Validity.
But we'll start with the two main types today, Internal Validity and
External Validity.
0:26
Internal Validity captures whether the study can demonstrate
that the independent variable causes the dependent variable.
Now, there are two main questions to consider when
looking at Internal Validity.
0:50
So let's consider the example of true grit.
Can we say that it's teacher's grit that leads to their effectiveness?
Or, could it be the other way around?
So for example,
it is possible that teachers who do better actually develop more grit over time.
That's a plausible relationship.
Because the better you do,
the more likely you might be to want to stick with something over time.
However, this isn't really an issue in our analysis because we collected data on
teacher's grit from their resume prior to them beginning teaching.
So we do know that their grit preceded their effectiveness.
1:34
Remember those variables we referred to in the last video as controlled or
lurking variables.
In the example of true grit, we did take some of those into account.
So we looked at things like, teacher demographics, teacher leadership
information available at the time of hire, and what the teacher actually taught.
However, there are other variables that could explain the relationship between
grit and effectiveness that we didn't measure.
So maybe it's not grit that causes teachers to be more effective,
maybe it's something else like teacher optimism, for example.
That's related to grit but actually explains differences in effectiveness.
Whenever you're reading a study think about other explanations that
could have explained the result that the author didn't take into account.
That's internal validity.
The second type of validity to consider is external validity.
External validity captures whether the results will translate to
other populations.
So for example, we need to consider if and
how our results will translate to different settings,
different groups of people, different places, and different times.
Take the example of true grit where we studied a specific population of teachers.
They were in their first two years in teaching in urban settings.
Had relatively high SAT scores and DPA's.
And had support from a teacher training organization throughout their
first two years.
So we have to ask ourselves if these results would translate to other settings.
Would grit predict defectiveness and
retention among more experienced teachers in non urban settings.
When teachers came from less selective colleges.
Now, whenever you're reading a study, think about the extent to
which the results will generalize to other populations.
3:16
That's external validity.
This is the final video introducing key concepts for this week.
There are few other components to this week, though.
Researcher videos,
featuring Angela Duckworth applying these concepts to her work.
3:30
Practitioner videos featuring practitioners applying these concepts to
their work, a short quiz on the key concepts we've covered this week, and
two optional extension activities.
A peer review activity, where you submit a hypothesis that you are interested in.
And have peers review that hypothesis based on the criteria for
effective hypothesis.
And provide feedback.
And an advanced activity where you review true grit or
another positivity psychology article of your choosing.
And discuss the hypothesis being tested.
And provide more detail in critique on the internal and
external validity of the study.
Next week we'll turn to data collection and measurement.
Claire Robertson-Kraft, Ph.D.Director, ImpactED
Angela Duckworth, Ph.D.Christopher H. Browne Distinguished Professor of Psychology
