Well, I watched the video, and I think Gov. Perry made a good point about "are we getting a return on our $ invested in abstinence education", and using steroid testing as a comparison, as an example of a program which may only catch a few trespassers and avert a limited number of kids from doing it, but is still considered a success if the state has a return on the investment (fewer people doing it means fewer people needing medical care for the health problems caused by steroid use).
Gov. Perry was using the correct logic, because you have to look at the difference caused by your method. For a rigorous scientific study, you would have to take two similar student groups in a Texas neighborhood, with the only difference among them being that one group would receive abstinence education, and the other group would not. Then, if there would be a difference between the pregnancy rates among these two groups, the size of that difference would tell us how much impact the abstinence education had.
I will use a counter-example to illustrate how simply saying that "Texas has the third-largest rate of teen pregnancy, therefore abstinence education doesn't work" does not constitute a rigorous scientific analysis and how it can lead to false conclusions. Here's the example:
"The USA has one of the largest mortality rates due to breast cancer in the world. A larger percentage of women die due to breast cancer in the USA than in poorly developed equatorial Africa. Therefore, breast cancer screenings, treatments and drugs developed in the USA do not work. Any money invested in breast cancer research and development of preventive screening, treatments, and drugs in the USA is a waste of money."
The above example is an analogy to the statement that "Texas has one of the highest teen pregnancy rates in the USA, therefore abstinence education as done in Texas is a waste of money".
In reality, the USA has some of the best screenings, treatments and drugs for breast cancer, and they do make a positive difference. But to show that difference, you have to compare the outcomes between two similar groups of women in the USA, one group of which has received no screenings and treatment for patients with breast cancer (control group), while the other group received state of the art preventive screenings, and treatments and drugs for those diagnosed with cancer. Simply comparing how many women die of breast cancer in the USA versus how many women die of cancer in equatorial Africa doesn't work, because breast cancer in poorly developed equatorial Africa is a rare occurrence to begin with (almost 10 times rarer than in the USA).
In my opinion, Gov. Perry has demonstrated a good logical grasp of how success of public health/social health programs is measured, while the journalist interviewing him didn't approach the question as a knowledgeable person would. In addition, or should I say foremost, Gov. Perry managed to uphold correct ethical values. Based on this video, Gov. Perry did not look like some scientifically and logically challenged simpleton to me, which is apparently the point that the huffingtonpost article was trying to make. It's more like the journalist interviewing him was too simplistic and un-scientific in his approach to the question.