Skip to content

Associated Press Propaganda: What the AP Survey Really Shows

August 18, 2013

On August 17, 2013, the Associated Press (AP) announced that a survey they conducted in June and July 2013 found that not only do parents really like standardized tests– they approve of the high-stakes usage of such tests and believe that the number of standardized tests administered is “about right”:

Often criticized as too prescriptive and all-consuming, standardized tests have support among parents, who view them as a useful way to measure both students’ and schools’ performances, according to an Associated Press-NORC Center for Public Affairs Research poll. Most parents also say their own children are given about the right number of standardized tests, according to the AP-NORC poll.

The release of this propaganda is certainly strategic timing given the recent New York debacle of Mayor Bloomberg’s aligning Common Core assessments to the National Assessment of Educational Progress (NAEP) and flunking most New York schoolchildren by design.

It is also interesting to note that at the same time the AP was conducting its survey, the Texas legislature finally heeded to parent pressure to reduce the annual number of standardized tests in Texas high schools from 15 to 5.

Nevertheless, according to AP, parents are fine with both the frequency and usage of standardized tests.

The AP report is a carefully crafted lie designed to promote an educational system that is high-stakes, standardized-test dependent at the expense of authentic teaching and learning.

Let us first consider who is purchasing this AP survey: The Joyce Foundation. Here is how the AP report spins Joyce Foundation involvement in their “research”:

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. [Emphasis added.]

The Joyce Foundation wants to judge teachers using standardized tests. On the tabulated survey results, the wording for Joyce Foundation “sponsorship” is more to the point:

Conducted and funded by The Associated Press-NORC Center for Public Affairs Research With major funding from the Joyce Foundation [Emphasis added.]

President Obama belonged to the Joyce Foundation board of directors from 1994 to 2002. Obama made it a requirement that states seeking Race to the Top funding must agree to evaluate teachers using student standardized test scores.

So here we have a “study” that finds parents in favor of what Obama wants, a study that just happens to have “major funding” from a reform organization to which Obama has close ties.

Let’s examine what the AP survey actually reveals. The survey is lengthy, so I will abbreviate results in this post. However, the complete survey is here.

This survey was of “1,025 adult parents of children enrolled in grades K-12 during the 2012-2013 school year.”  For parents of more than one child, responses are based upon the oldest child’s school experience.

76% believe that their child’s school rates “excellent” (37%) or “good” (40%).

57% believe that the school is doing an “excellent” (18%) or “good” (39%) job preparing students for college– up from 48% “excellent/’good” on the AP survey in 2010.

The highest ranking factor that contributes to the quality of a student’s education (“extremely important”) was “the amount of parental involvement in the child’s education” (65%). Second was “the quality of the teachers” (61%).

For some reason, the researchers used only half of the sample for next set of questions. (The overall sample was divided into two and labeled “A” and “B,” with some questions answered using “half sample A” and others, “half sample B.”) The questions pertain to “problems facing schools today,” including but not limited to “low expectations for student achievement,” “inequality of funding,” “quality instruction by teachers,”  “students not spending enough time in school,” and “low test scores.” The results of this section are flat. The highest category was that 33% of “half sample A” thought that “not enough opportunity for physical activity and sports” was “not too serious” of a problem in school.  The highest ranking in the “extremely serious” category” was 20%, for “getting and keeping good teachers” (“half sample A”).

From 2010 to 2012, the percentage of parents who believed that the education their child is receiving is “much better” than their own education increased from 31% to 38%.

So far, these results are not panning out for reformers wishing to “innovate disruptively.”

Now we come to a loaded question choice.  In response to “determining the quality and performance of a school,” parents are offered the choice, “information about teachers’ ability to improve student outcomes.” In reformerspeak, this is the all-too-familiar wording for “grading teachers using students’ standardized test scores.” Notice, however, that the response choice mentions nothing about use of standardized tests; parents are simply offered the nondescript term, “information.” 38% rated this “information” as “extremely helpful.” I do not believe that respondents were thinking of value added teacher evaluation, and here’s why:  The two items mentioning student tests, “average student test scores” and “changes in student test scores over time,” were near the bottom of the list of what parents deemed “extremely useful” to determining quality and performance of a school” (both at 25%). Information on their children’s standardized tests was not as “extremely important” to parents as was “the school’s safety and security record” (37%), “information about teachers’ academic and training backgrounds,” (33%), and “information on the school’s budget and spending” (29%).

The lowest item on the list of “extremely useful” in determining school quality was “student dropout rates” (24%), another area of reformer numeric fixation and manipulation.

How do parents gather information on school quality? Mostly by asking other parents (83%) and visiting the school’s website (65%). Overall, 82% of parents rate the quality of their child’s teachers as either excellent (42%) or good (40%). When asked to expand upon what makes the teacher “excellent,” the highest ranking response was “attentive to student’s needs” (35%). “General teacher performance/effectiveness” ranked among the lowest, at only 2%.

All four of the highest ranking responses concerned quality of human interaction (the remaining three were “presentation style” [22%], “good communication” [12%], and “positive attitude and personality” [10%]).

The same four qualities of human interaction responses were the top four reasons that parents also rated some teachers as “poor.” Only 6% said the teacher was “poor” due to “general teacher performance/ effectiveness.”

Parents want teachers who care for and interact well with their children.  This value is reiterated in parents’ responses to the following question:

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%).

As for teacher pay, in both 2010 and 2012, the largest category for parent perceptions of teacher pay was that teachers are paid “far too little” (32% in 2010; 38% in 2012).

Now for some AP reporting trickery. Here is an excerpt from the “pro-testing” AP article:

They’d (parents would) like to see student performance on statewide exams used in evaluating teachers….

And now, for the survey questions. The researchers split the sample into two halves in order to offer two versions of a question regarding judgments on teacher salary.

The first question offers three choices: teacher pay based solely on statewide tests; pay based on both statewide tests and administrative classroom observations, and pay based solely on administrative classroom observations.  The highest category was the “both” category in both 2010 and 2012 (49% in 2010; 50% in 2012). I am not surprised by this result; it would be helpful to know why parents chose as they did. I suspect that it seems “fair” to incorporate both means in determining pay.

Neither teacher seniority nor education experience were choices in this version of the question, which makes the result easier for reformers to present as parents “favoring” use of standardized tests to evaluate teachers.

Respondents were also not informed that teachers could possibly be fired based upon the evaluation ratings.

In other words, parents were not apprised of the “high stakes” mentioned in the title of the AP article.

In the second version of the question, parents are given more choices for determining teacher salary; they are also given the opportunity to register varied levels of agreement for use of each.  In this version, parents were hesitant to use the “extremely important” category, instead leaning on the second, “very important” category. What parents believed was “very important” in determining teacher salary was “classroom observations by local school officials” (43%); “the type of training or advanced degrees obtained by the teacher” (40%); “years of teaching experience” (38%); “changes in student test scores over time” (34%), and “input from parents” (30%).

As was true of the first version, the second version of the question focuses only on teacher salary and not teacher job security.

No mention of “high stakes.”

There is more to the AP release statement than I copied previously. Here is the full statement:

They’d (parents would) like to see student performance on statewide exams used in evaluating teachers, and almost three-quarters said they favored changes that would make it easier for schools to fire poorly performing teachers.

The AP article connects the ideas of “teacher evaluation” and “teacher firing”; however, there is no reason for parents completing this survey to believe that “evaluation” heavily reliant upon “standardized test scores” could lead to the “firing” of teachers by design.

The AP survey question on teacher firing is worded in such a way that most would naturally agree:

Would you favor, oppose, or neither favor nor oppose making it easier for school districts to fire teachers for poor performance?

Hey, your child has a bad teacher. Would you like to make it easy to fire the bad teacher?

My surprise here was that anyone responded, “strongly oppose” (6%, in both 2010 and 2012). Then I realized that some parents in this sample are also teachers, and they likely know the game being played in the name of reform.

The AP article also includes this statement, a clear jump in logic based upon the actual survey questions and results:

The polling results are good news for states looking to implement increased accountability standards and for those who want to hold teachers responsible for students’ slipping standing against other countries’ scores. [Emphasis added.]

Over one third of parents in this survey believe that their children are receiving a “much better” education than they did. It is the largest response category for the question. The number has increased from 2010 to 2012.

The survey does not mention international testing.

Now comes the time for this Joyce-Foundation-funded survey to make good on its endorsement of teacher evaluation via standardized test scores. Even though the researchers have already asked questions in which “standardized testing” was implied (e.g., via use of the term “statewide tests”), the researchers wait until well into the survey to define the term. Their definition is rather weak:

Standardized tests are used to evaluate students using a consistent test so that each student is scored in the same way.

There is no mention of testing companies, no mention of the usage of such tests, no mention of the time it takes to administer these test, no mention of the cost to the district, and certainly no mention or defining of the term “high stakes testing.” Thus, when parents were asked if their child takes too many standardized tests, they relied upon the general definition offered and likely responded according to “tests in general.” Many of these parents did not think of state or national tests– and surely not of international tests. I base my judgment on the fact that 46% of parents believed that “the local school district should be responsible for the subject areas covered in these standardized tests.”

The parents were hearing “standardized tests” and almost half were thinking these should be “locally controlled.”

The question regarding the use of standardized tests to evaluate teachers includes a “normalizing” statement at the outset. This statement is clearly present in order to bias the respondent into believing that using standardized tests is “okay” for any and all of the choices that follow:

All public schools give their students standardized tests from time to time. Do you think standardized tests should or should not be used for the following purposes? [Emphasis added.]

The question biases the response that standardized tests “should be used” for all purposes presented, including “to rank or rate schools” and “to evaluate teacher quality.”

There is no mention of using standardized test results to close schools or to fire teachers. This survey obviously skirts the “high stakes” conequences associated with reformer standardized test usage.

As for Common Core: The questions are clearly biased in favor:

Most states have adopted the Common Core State Standards in mathematics and English language arts. The objective of the Common Core is to provide consistent, clear standards across all states for students in grades K-12. [Emphasis added.]

What parents would not want “clear, consistent standards” taught in their child’s school?

Even though half of the parents have heard “little or nothing at all” of Common Core, the researchers use this opportunity to plant a seed of acceptance for Common Core by offering their biased definition to an uninformed audience.


The offering of a slanted definition of Common Core is the deception behind this AP article statement:

Still, when given a brief description of what the standards do, about half of parents say educational quality will improve once the standards are implemented, 11 percent think it will get worse, and 27 percent say they’ll have no effect. [Emphasis added.]

In sum, the AP article does not reflect the declared purpose of the survey as evidenced by the survey questions. The parents completing the AP survey were not instructed in the use of the term “high stakes,” including the potential, serious outcomes of high stakes testing.  They were also not informed of the high-stakes-testing requirements associated with Common Core.

If one considers the survey results separate from the AP article, one sees parents who believe their children are receiving a better education than they did from excellent-yet-underpaid teachers who care about their students.  I dare the Joyce-Foundation-funded AP to print that info.


  1. Linda permalink

    Thank you Mercedes for investigating and for keeping all of us informed. You are a treasure.

  2. Thank you!! This is just what we needed!! Linda is right – you are a treasure!!

  3. Laura H. Chapman permalink

    Remarkable analysis. The AP survey is known as a “push survey.” it is designed to push an agenda. Bellweather is a notorious provider of push surveys for anyone who will foot the bill. Now we know that AP does the same if someone pays the bill. The liinkage of Obama with the Joyce Foundation is new to me, but so was the link from McKinsey & Co to Arne Duncan and David Coleman, dubbed the architect of the Common Core State Standards, now CEO of the College Board. You did a great job of “thinking like a parent” faced with this survey.

  4. The anti-social media cannot be trusted on this score since they have become competitors for what used to be public education dollars.

  5. patriciahale permalink

    Nice investigative work. You can’t trust anything you read nowadays, without looking deeper.

  6. David B. Cohen permalink

    Reblogged this on InterACT and commented:
    Mercedes Schneider digs into the details of this recent poll and finds, not surprisingly, that the questions skew the results by providing a favorable view of the topic ahead of the question, and by constructing responses options that encourage a certain result. Meanwhile, reports and press releases seem uninterested in some other findings from the survey, results showing that parents “believe their children are receiving a better education than they did from excellent-yet-underpaid teachers who care about their students.”

  7. bernie1815 permalink

    For what its worth, I think you have done a good job pulling apart the AP report. I haven’t spent a lot of time on the results since the survey is really capturing attitudes that may or may not be informed – a point that you rightly stress.

    Based on conducting surveys over the last 30 years, I believe the survey is way too long. In overly long surveys respondents begin to chose answers that reflect the least amount of psychological effort. Some careful quality control of responses are needed. If there is not some significant attrition in the sample size then I would be very surprised.

    The survey sample includes a surprising 20% who are in a household with at least one teacher. This may explain the aberrant finding about what to do about a bad teacher that you caught.

    Responses aggregate parents of Elementary and HS students. There appears to be no separate category for MS students.

    The survey was conducted outside of the school year during the summer. The earlier 2010 survey used as a comparison was conducted during the beginning of the school year in September. Neither strikes me as optimal.

  8. Ed Week has now picked up the story you debunked.

  9. susannunes permalink

    It’s an excellent analysis, but all one has to do is look at who sponsored it and not even have to analyze it to know it is a bunch of bunk.

    It’s like reading a “study” conducted by the Cato Institute about Social Security. You already know the conclusion without even reading it because the Cato Institute, like the Joyce Foundation, is a propaganda organization masquerading as a think tank.

  10. Margaret permalink

    Thanks for this informative analysis. As a side note: I tried twice to sign up to get email updates for the blog. Both times, I got an email message with a link, but when I clicked the link it did not work to sign me up…..

    • Margaret, try using a different computer. My computer at home has an issue with the safety settings, and it blocks such links. Sometimes it will allow me to copy the link and paste into another window.

  11. The AP has published a follow-up of sorts. The article focuses on the differences in the responses of parents of different races. It doesn’t jump to conclusions as badly as the first article, but it is poorly organized and seems to cherry pick which questions to discuss again.

    The authors choose a bunch of semi-connected questions to throw out there, but don’t do any analysis and then choose random parents to interview to frame the data. There’s two guys with negative views toward education (US schools worse than the UK and teachers give too much homework because they’re too lazy to teach themselves) and the teacher worried schools aren’t rigorous enough. I think they misquote him via summary about money “not determining whether the students succeed or fail.” I think he just means that his affluent community has decently funded schools, while their summary of his words could be construed to support the reform “money doesn’t matter” meme.

    They mention the finding that parents are satisfied with their schools, but never follow up on any of the positive findings.

  12. Anyone else noticing that a lot of the numbers don’t even add up to 100% ? ‘About half’ implies the number is less than 50%, then 11%, then 27%…even if you credited it at the full 50%, you add the other 38% and you get 88%, not 100%. Maybe the AP is the one who needs to be taking standardized tests…in 1st grade mathematics.

    • Some respondents could be dropping out of the survey. It is a long survey.
      This is just one possibility– I don’t have the numbers in front of me.

      • bernie1815 permalink

        The AP is reporting % and does not appear to tabulate actual numbers of responses for all questions – so we do not know whether respondents are dropping out of the survey.

        I think, based on a quick review, that the “Don’t Know” categories accounts for the fact that the numbers do not add up to 100%.

  13. Beth Flynn permalink

    I think you should send a link to your blog to the AP. I sent them an e-mail yesterday after reading the article and reviewing the survey (which does not have any information broken out by race but they gave information about it in the article). They have “values and principles” which they, obviously, didn’t follow when writing the article: Here is their e-mail address: Maybe if enough people call them on their dishonest reporting, they’ll get back to actually reporting the news. I sent them links to conferences for people interested in investing in education and told them THAT is the story they should be investigating.

  14. Cosmic Tinker permalink

    Here’s another one for you, Mercedes, enlightening info about the CREDO study:

    • bernie1815 permalink

      Cosmic Tinker:
      I agree. If the Louisiana DOE has no acceptable rationale for denying Research on Reforms access to the same student data as used for the CREDO report then the report should be viewed with skepticism. Transparency and replication are essential.

      Of course, Research on Reform’s hypothesis may be proved wrong but they clearly should have the opportunity to test it.

Trackbacks & Pingbacks

  1. Associated Press Propaganda: What the AP Survey Really Shows | deutsch29 ← NPE News Briefs
  2. How the AP Turned a Survey into Propaganda | Diane Ravitch's blog
  3. Mercedes Schneider: What You Should Know About the Deceptive AP Report on Standardized Testing | Political Ration
  4. Mercedes Schneider: What You Should Know About the Deceptive AP Report on Standardized Testing | Both Sides Clash
  5. Schneider to David Brooks: Common Core Does Not Encourage Opinion Writing Like Yours (UPDATE) | Diane Ravitch's blog

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s