professornana (professornana) wrote,

  • Location:
  • Mood:

Massage, anyone?

I took quite a few stat classes when I was working on my doctorate. I actually enjoyed them (I know, weird, right?. Part of the reason was because I had a lovely study group which included my pal Kylene Beers. We told ourselves going in to the first course that we could get a "C" and still be doctoral candidates. It took the pressure off, and that proved to be one good step toward doing well in the course. Another plus was the incredible professor who took us slowly through each and every element to ensure we understood what we were learning. But, most of all, I loved these courses because they taught me how to read between the lines where numbers are concerned, to tell the difference between numbers one might rely upon and number that might be massaged to prove whatever point someone wanted to make.

So, when I see a report on a poll or research, I often look closely to see the real story behind the numbers. Such was the case this past week when Arne Duncan lauded a new poll that showed most Americans in favor of CCSS and its accountability measures. On the surface, this poll looked legitimate. You can read the AP story here: But I knew there had to be more to the story. I kept reading and got to the final paragraphs: "The survey was sponsored by the Joyce Foundation, which works to promote policies that improve the quality of teachers, including the development of new teacher evaluation systems, enhance early reading reforms and encourage innovation in public schools. The AP-NORC Center for Public Affairs Research survey was conducted June 21 through July 22, 2013. The nationally representative poll, conducted by NORC at the University of Chicago, involved landline and cellphone interviews in English or Spanish with 1,025 parents of children who completed grades K through 12 in the last school year. Results for the full sample have a margin of sampling error of plus or minus 4.1 percentage points; it is larger for subgroups." Okay, this was becoming a bit more clear. Small sample. Not much information about the sponsoring organization. No mention of reliability and validity. I wondered what else I was not seeing? I suspected a deep massage.

And then, someone on Twitter provided me the link I sought: Read this post as it uncovers some of the flaws (if not downright lies) of the poll. Take a look at what was NOT reported. Items like this one: "If you could choose your child’s teacher, what would be the most important factor for you in choosing the best possible teacher? Top choice: Passionate about teaching (21%). Second choice: Caring toward your child (12%). Third choice: Evidence that the teacher’s students are learning (9%)".
It is enlightening to note that parents still want the human element to be the deciding factor in what makes for a good teacher.

So, when you see numbers being cited, ask yourself what they really mean. As we collect data (and under CCSS, data is king, the coin of realm to mix metaphors), ask what it really tells you about the KID behind the numbers. When College Girl came home with her ITBS scores and showed me she scored 11.0 as a 5th grader, we had a talk about what that meant. Someone at school suggested it meant she read at the 11th grade level. That's not quite true. We talked about the implications on this test and the never-ending series of tests for AR and state assessments and what those numbers really revealed. But how many kids go home where the parent is NOT in education or research or a field where this information is common knowledge? There are bumper stickers proudly proclaiming that kids "aced" the state-mandated tests. Um, not quite accurate. Passing scores on the old assessments (and most likely the new ones here in Texas) were set AFTER the test was administered. A school's "grade" took into account the number of free and reduced lunch students, attendance, and other factors. I know that. I am not impressed with the exemplary status of a school that uses this as a measurement.

You want to impress me? Talk to me about how your students love to read, how they read for pleasure, how they talk about books, how they write for outside audiences, how they participate in extracurricular activities. Tell me about what you do for the WHOLE CHILD to borrow a phrase from the past. Tell me without all the numbers.
Tags: data, lies, numbers, polls, stats
  • Post a new comment


    default userpic

    Your reply will be screened

    Your IP address will be recorded 

    When you submit the form an invisible reCAPTCHA check will be performed.
    You must follow the Privacy Policy and Google Terms of use.