New survey finds surprising opinions about tests

A new national poll offers a basis beyond anecdote to report what parents, children and educators think about tests.

IMG_20160216_112740

Feb. 16, 2016, Atlanta — A protest against testing during a rally at the Georgia Capitol. TY TAGAMI/AJC

Despite a public outcry, only about 15 percent of parents feel strongly enough to opt their children out of the tests and most students don’t feel over-tested, according to this study. It’s from Gallup and the Northwest Evaluation Association, a not-for-profit that markets the Measures of Academic Progress test to schools.

The pollsters found that opinions about tests diverged based on income: wealthier parents are less likely than the poor to agree that tests improve learning. Also, administrators at schools with data coaches to help interpret test results say teaching has improved, and low-income schools more frequently report having a data coach.

Another key finding: there is a lot of confusion about the different types of tests, and there are many, each offering different kinds of information. Summative tests, for instance, measure what students have learned at the end of the year, and are used to judge teachers. Interim tests identify students’ strengths and weaknesses and track their growth. Formative tests help teachers figure out what’s working, or what isn’t, with the way they are teaching.

The people at NWEA fear that the frustration over “accountability” tests used to judge teachers will turn the public against all tests. So they did this study — actually the third one since 2012 — to bring some clarity to the subject.

“We feel there’s a lot of misunderstanding around assessment perceptions, and that gets in the way of education,” said Kelly Goodrich, the vice president for policy and advocacy at NWEA.

The group may have a vested interest, but its survey is based on phone interviews with about a thousand students and with a similar number of parents and teachers. It has a margin of sampling error in the 4 percentage point range.

You can judge the survey for yourself here.

Reader Comments 0

12 comments
Starik
Starik

Once again, a lack of real standards embarrasses the teaching profession.  WSBTV had another piece on this charming lady, graduate of hbcus and internet "universities."  Does anyone risk swallowing their tongue addressing this lady as "Dr.?" 


Test beginning teachers, with a real exam, one that can be failed.  Quit paying salary bonuses for fake degrees. Fix this.

Starik
Starik

I said a bad word, quoting an "educator" and got blocked. DeKalb did it again, a "Dr." at Chamblee HS - one of their few decent high schools - getting away with being stupid. 


Dr. of what? From which respected university? In what subject?


Maybe Ms. Downey will call WSBTV and pick up on this one.

Lauren Ess
Lauren Ess

This caught my attention and left me skeptical and curious after reading the original poll and (original) accompanying article sponsored/written by the NWEA. Is it written very broadly/loosely in order to support assessment testing? The report includes several very different tests together: classroom/teacher developed assessments, formative assessments, diagnostic assessments, interim/growth assessments, performance tasks, summative assessments, accountability assessments, system-level assessments and international benchmark assessments. The tests that the principles, teachers, and students value most in this report (from what I gathered) are on a micro or classroom level -- the formative tests, for instance. Within the report, it ranks the accountability assessments as the least desirable (pg 4 at 37% -- See pg 48 for definitions). The accountability assessments are those like Common Core tests issued by the state. The formative tests are things like quizzes and tests issued by teachers to reinforce learning or understand deficits, is my understanding. So with that said, is this information that surprising? I'm still seeing that principals, parents, and students don't necessarily want or agree with assessment testing (Again, 37% versus high 90% for teacher-initiated quizzes and tests). Also, since the poll and report details low-income families as being those most opposed to testing, I'm curious which demographic(s) they polled since they clearly highlighted socioeconomics. There is a great deal of hard/raw data left out of here. Lastly, one thing I do agree with is that most people don't know how test results are used. Since 70+% of teachers don't feel comfortable interpreting or scoring tests, and most schools don't have "data coaches" to teach them, is testing really that effective in learning and teaching on a micro level or is it just about gaining additional government funding for higher test scores? And don't the poll results read that educators, parents, and students prefer, do better on, and value classroom sponsored quizzes and tests instead of the marathon state-issued assessment tests?

jerryeads
jerryeads

Starik, we already do at entry, and the state's licensing agency has ramped up the difficulty over the years. AND they've added some performance evaluation along with it. It WOULD be interesting to require re-certification examination every decade or so, just like MDs. HOWEVER, knowledge of the subject does not mean someone can teach it. Witness the far too common college faculty who are quite expert in their content but are just gawdawful at teaching it, and could care less. Most K-12 teachers work very hard to get better at what they DO (not just know) every day, but given that teaching is one of the most incredibly complex skillsets on the planet (try it sometime with a roomful of 30 10-year-olds), it's a pretty hard thing to measure either meaningfully (validity) or accurately (reliability).

xz, There are such things as "good" tests. They're well-made, and are used for the purpose for which they're intended. Unfortunately, not only are many low-bid state-required minimum competency tests egregiously poorly made, they're built ONLY to determine if a kid has performed at least at a very specific level on a very specific topic - - and then we use them for all sorts of things (like teacher evaluation) for which they were never designed and, in fact, cannot do meaningfully OR accurately.

Starik
Starik

@jerryeads I respect good teachers who know their subject and have good classroom skills. I believe their pay should be doubled, at least. Why should teachers like you and the others who post here be paid miserable salaries because the standards for entering the profession are so low?  You've seen the surveys (no I don't have a link) showing that large numbers of teachers are graduates of the very worst colleges and were at the bottom of the class there. As a student, then parent, then grandparent I've seen it. Waterloo is not in Russia. Rommel was not an Admiral. The KKK was not an admirable organization even after the "War Between the States" when the Yankees let the [African Americans] run the government. Russia defeated the Germans in WW2, mostly. Our war was with Japan, mostly.  Men do not have one fewer rib than women. We evolved. We did not mix with dinosaurs. "I be" and "I is" are not correct English. Teachers demand respect, and if a kid insists that Waterloo is in Belgium it's considered disrespectful and punishment is required.  Coaches should usually just teach PE, and shouldn't be paid better than real educators or be promoted to Principal unless they can teach the subject. 


Testing the kids to evaluate the teachers, I agree, is unfair because there is a vast difference - vast - between what the kids bring to the classroom in the morning and go home to in the afternoon. 


The 1950s were not paradise in the schools, when teachers could beat the kids with the classroom paddle (some enjoyed it a little too much) and we were segregated by race.  If you teach in Fulton County you are surrounded, largely, by good colleagues. DeKalb? Atlanta?  Not so much...  I see that the convicted teachers in the cheating scandal are being allowed to teach school again. 


I understand that there are technical skills in presenting material in the classroom, and that teaching kids math or reading to beginners is different from teaching middle or high school. 


I agree that judging teachers by testing students is unreliable. There's a huge difference between what kids bring to the classroom in the morning and go home to in the afternoon.


Still, explain to me how a teacher can teach a subject they don't know?  How to handle questions from students?  How can kids learn to speak standard English when neither parents nor teachers can?  What does this do to the kids' future?

Starik
Starik

Why not test the teachers, directly, on their knowledge of the subject they're teaching?

bu22
bu22

Most people thinks tests are good.  That's hardly a surprise unless you only talk to educators.  As Redweather suggests, there is a strong feeling that too much time is devoted to testing, but that doesn't mean people think no time should be devoted to it.  As for the teacher evaluation part, that is a much bigger issue for those immediately impacted, teachers, than for the rest of the public.  The rest of the public is focusing on the children, not the adults.

Tom Green
Tom Green

Take classroom based assessments out of the survey and rest of the data is virtually useless for driving classroom instruction and improving student learning. With classroom instruction being driven by benchmark testing, grade level planning, and the need to finish all instruction at least a week ahead of the state's accountability test so that you can review the past 28 weeks of learning (college students are only accountable for 16 weeks), there is extremely little time for reteaching missed concepts. As for analyzing all of that data to personally tailor individual instruction and to personally contact parents to explain results? Come teach for a week and and show me when you would make that happen.

Legong
Legong

Maureen, you suggest the group commissioning this research may have a vested interest. But when have you ever questioned the obvious vested interests of the teachers' union bosses behind "opt-out" and similar fronts opposed to accountability and parental choice?

redweather
redweather

First, this is really about four issues instead of one: (1) testing, (2) how much time is devoted to testing, (3) how much weight should be given to student performance on tests, and (4) how much weight should be given to student performance on tests when evaluating teacher performance.

But there is a related issue, which I will call the elephant in the room. What I see at the collegiate level is that a growing number of students have under-developed reading skills, and this adversely affects their performance on tests. If I'm seeing this at my level, then I know this is a factor at the K-12 level. A test, whether it be formative, interim, or summative may not accurately identify a student's subject knowledge if that student's reading skills are inadequate.

In other words, get them offline and un-phoned if you want your children to succeed in school. It really is that simple.



Starik
Starik

@redweather You can't turn back the clock. When they're on the computer they're reading something.  When they're texting they're sort of writing.  Perhaps a bigger problem is gaming, and judging from the blog posts infesting the AJC blogs, facebook, which seems to be full of idiots  and illiterates.