Harvard, SurveyMonkey Offer Tool to Weigh Parent Engagement
Harvard, SurveyMonkey serve up template
A new survey tool that school districts and parent-teacher organizations can use to measure the quality of parent-school relationships has been created by the Harvard Graduate School of Education and released by SurveyMonkey, a Palo Alto, Calif., company, for widespread use by schools, districts, and parent groups.
The 71-item "question bank" covers seven areas of family engagement—from how much help students receive at home to how confident parents are in supporting their child's schooling. Districts can adapt the survey to suit their individual needs, and parents responding to it can do so online or on paper.
After introducing the survey last May, SurveyMonkey is working directly with 12 early-adopting districts on its deployment, and others are finding out about it online or via connections with Harvard. Educators and parent organizations can begin using the survey now without enlisting SurveyMonkey's help.
Karen L. Mapp, a lecturer on education at Harvard and the director of the education policy and management master's program there, provided her expertise in parent engagement to the survey's development.
"Districts are really trying to get parents more engaged in supporting their children's learning; there's more of an academic focus now on engagement efforts," she said. "Districts are saying, 'We want to change the paradigm on parent engagement, but we just don't have the tools to measure that.' "
At the same time, school districts that won Race to the Top federal grants pledged to measure school climate and collect data that includes information about how parents perceive school effectiveness.
Ms. Mapp, who was a family- and community-engagement official with the Boston district in 2003-04, said the questions in the tool can be used with that goal in mind as well.
The Harvard Graduate School of Education developed and field-tested a 71-question survey that schools can use via the SurveyMonkey website to solicit parents’ feedback about their children’s educational experience. The survey, divided into seven topical sections, can be offered in its entirety to a school community, or schools can select questions from the most relevant categories and add their own.
"When I was deputy superintendent, we didn't have anything like this. I was turning to my data colleagues saying, 'Can you develop a quick survey?' " recalled Ms. Mapp.
Schools and parent-teacher organizations will be able to use the results from the new parent survey to direct their parent-engagement efforts more effectively, according to Ms. Mapp. "It will give parents and schools more decisionmaking power about how to more effectively help their kids excel," she said.
Tens of thousands of schools have been using SurveyMonkey to poll their parent communities for more than a decade, according to Bennett Porter, the vice president of marketing communications for the company.
But not all questions are created equal. Many of the queries crafted by school officials missed aspects of family-school relationships that play key roles in students' performance.
"Given the number of people who are asking questions using SurveyMonkey, we noticed that there's a big opportunity to help people ask common questions in ways that are methodologically sound. … Let's take these questions and write them in a way that will minimize bias and the risk of getting bad data, because in the end, you're using the answers/insights gleaned from a survey to drive important decisions you're making for your school, so you want to be really careful to have good data," said Will Aldrich, SurveyMonkey's vice president of product.
SurveyMonkey reached out to Harvard to identify which questions would produce accurate information that districts could rely on as meaningful for their assessment and decisionmaking about parent involvement. The company funded a research grant for the Harvard Graduate School of Education to underwrite the research and development of the questions. Mr. Aldrich declined to disclose the amount of the funding.
Hunter Gehlbach, an associate professor in the school of education, led the team that developed the survey between September 2011 and May 2012—a very aggressive turnaround for academic survey designers, he said.
Arriving at the "right questions" has been an arduous process, he said. "Everyone's taken surveys, and they think it's real simple, but there's a huge amount of science behind it," said Mr. Gehlbach, who explained that four doctoral students, Sofia Bahena, Lauren Capotosto, James Noonan, and Beth Schueler followed a rigorous six-step process under his guidance.
Starting with a review of the academic literature, interviews, and focus groups with parents, the researchers developed a series of questions around seven "constructs" or topic areas.
"Even with all the advance research, you don't want to have supreme confidence that you've designed the right items," Mr. Gehlbach said. "So we went to the experts, the academics and practitioners. We made some changes based on their feedback, and took the [questions] to parents, to make sure they not only understood the wording of the items, but that they resonated with the language."
That exercise proved to be eye-opening. "For example, in a topic like 'parental support,' one thing the literature picks up on is 'parents helping with homework.' ... In focus groups, the parents actually tell us, 'No, no, no, that's not so right. We don't help our children with specific homework questions; we focus on getting our children to understand the overall content they are learning in class.' "
As a result, no questions using the word "homework" appear in the survey. Instead, questions focus on helping a child understand the content he or she is learning in school.
"The literature is incredibly helpful, but you can't just take the academics' word for it," said Mr. Gehlbach. "You have to check in with the parents as well."
Another area where parents diverged from academic language was in the use of the term "student engagement." Parents had different interpretations of what that means, so the survey designers chose "classes that motivate children to learn," rather than mentioning "student engagement" anywhere.
In defining the answers—which academics call "response anchors" like "very satisfied," "somewhat satisfied," and "not at all satisfied"—Harvard researchers relied again on science and best practices to ensure that parents' answers would produce meaningful data.
Harvard's goal was to group the questions into discrete scales so that the survey would offer flexibility to users interested in a particular subject area. The seven constructs measured by the scales are:
• Parental support: How much help are students getting at home?
• Child behaviors: What habits have students developed that shape their success?
• Parent engagement: How engaged are parents in their child's schooling, and what potential barriers exist?
• Parent self-efficacy: How confident are parents in supporting their child's schooling?
• School climate: How do parents view their school regarding academic and social standards?
• School program fit: How well do a school's academic program, social climate, and organizational structure meet a student's needs?
• Parent roles and responsibilities: How do parents view their roles, as well as teachers' roles, in different aspects of their child's education?
The surveys will be useful in many scenarios, such as helping educators get a baseline measure of how parents perceive the school climate, determining what kind of learning-related behaviors students are engaged in at home, and identifying where the strengths and weaknesses are in a school's relationship with parents, Mr. Gehlbach said.
Schools and parent groups can access the Question Bank, a repository of questions anyone can use for free to ensure a survey has well-written questions. To ask more than 10 questions in a survey via the SurveyMonkey platform, organizations would be charged according to one of three plans, ranging in cost from $17 per month to $65 per month. The top level offers the most analytical and support options.
Mr. Aldrich said that working with the first cohort of early-adopting districts will give his company a good idea about what tactics work best to maximize response rates. In general, choosing the scales that are most relevant to a district will help, as will keeping the surveys on the shorter side. Parents can be invited to participate by email, with reminder emails sent to only those parents who did not respond.
"Everything about SurveyMonkey is intended to be self-serve and easy to get started," said Mr. Aldrich at SurveyMonkey, explaining that schools and organizations are under no obligation to talk to the company first. "The template is available now, and schools are using it now."
Recognizing the digital divide that exists for some respondents, he said the surveys can be completed using the paper-and-pencil method as well. The questions have also been translated into Spanish.
Offering a Resource
SurveyMonkey officials declined to name school districts that are using the survey. But Michael Sarbanes, the executive director of the office of engagement for the Baltimore public schools, said he is aware of it through his participation in the District Leaders Network on Family and Community Engagement, held at the Institute for Educational Leadership, a Washington-based nonprofit organization. He called the new survey a good resource.
"We have our eye on it. One of the critical pieces in [our] work is getting actionable information … that is real-time and practical enough that you can adjust what you're doing based on the input you're getting. I think this will be a helpful tool for that," Mr. Sarbanes said.
Vol. 32, Issue 17, Page 10
Get 10 free stories, e-newsletters, and more!
- Superintendent of Schools
- Regional School Unit 21, Kennebunk, ME
- Ridgefield Public Schools, Ridgefield, CT
- Principal, Niwot High School
- St. Vrain Valley School District, CO
- Administrative Assistant
- AERO HYDROLIC INC, Chicago Heights, IL
- Accomack County Public Schools, Accomac, VA