This week the primary teachers’ union NZEI released results from a poll intended to reveal people’s preferences for how public money is spent on education.
According to the NZEI, “less than 6 per cent of people think the government’s plan to establish new leadership roles for some principals and teachers is a good use of increased education funding”. The public seems to place a much greater emphasis on reducing class sizes (32 per cent) and ensuring all early childhood teachers are qualified (31 per cent).
Of course, it is worthy to explore what the public actually thinks about policy.
But I can’t help going back to questionnaire design 101 when considering the results. One rule of thumb sticks out: avoid loaded questions.
Respondents were asked, if the government were to increase public school funding, of a list of areas where that money could be spent, which two were the most important? One of the options was “paying $40,000 more a year to executive principals to oversee a group of 10 or so schools”, referring to the new leadership roles.
Other options included “reducing class sizes” and “ensuring all children get the best start in life with ECE [early childhood education] services with 100 per cent qualified and registered teachers”.
The problem with the latter option is that it is loaded with a cut and dried presumption that having fully qualified ECE teachers is the most effective policy lever for ensuring all children get the best start in life.
I am not familiar enough with the research to know whether this is true. But I am familiar enough with the basic traps of questionnaire design to say that it is hard to take the survey seriously. This criticism could be easily avoided by removing the potential for biased responses. Just “ensuring 100 per cent qualified and registered teachers” would do.
To further illustrate the point, what if the tone of the response option about the new leadership positions was different? If the response option had been worded “ensure we keep good teachers by offering an aspirational career pathway with more leadership positions for teachers”, the ticks in that box would surely be higher than 6 per cent.
How so? Well, another survey that captures the mood of New Zealand’s primary schools came out this week. The New Zealand Council for Education Research (NZCER) asked teachers, principals, trustees and parents about the major challenges facing their school. The top challenges identified by parents were funding (39 per cent), keeping good teachers (28 per cent), and large class sizes (24 per cent).
This survey would suggest that keeping good teachers is as important to parents as small class sizes. So, by loading a question with a reference to keeping good teachers this would likely bias the responses in its favour.
The NZEI represents the teachers who are faced with the realities of the classroom, and policy makers should take heed of their advice. But ensuring their research is not laden with bias would go a long way towards giving teachers a voice of credibility.
The pitfalls of survey design
9 May, 2014