Types of Questionnaires

A questionnaire can be administered in four basic ways. It can be a pencil-paper questionnaire (1) or online (2), conducted in the form of an interview (3), or over the phone (4).

Teachers use questionnaires a) standardized, which have fixed questions, a manual describing how to work with the questionnaire and sometimes the norms for the population; b) non-standardized, which, for example, a teacher prepares for their own needs.

Creation and Use of Questionnaires

If we use a ready-made professional (standardized) questionnaire, it is necessary to familiarize ourselves with the manual for its use, the limits of the tool, the procedures for its administration, and the method of processing and interpreting the results. We need to verify whether the questionnaire is applicable in our situation with our students in the given age cohort.

If we use a non-standardized questionnaire, we need to check whether it properly represents the measured construct and corresponding theory (e.g., class climate), whether the items are understandable, etc.

If we create a questionnaire ourselves, we need to rely on the rules of questionnaire construction, which are similar to the rules for constructing a knowledge test or interview questions. In many cases, we can simply conduct a structured conversation with the student on the given topic according to the questionnaire. The rules for creating a printed or online questionnaire were discussed in the course "Research in Educational Practice" in the fall semester of the second year.

- In principle, it is always necessary to start from our diagnostic question, diagnostic goal, assumption, or hypothesis and design a well-thought-out structure of the questionnaire accordingly – areas which we then fill with items. We usually base this on a theory about the issue we are diagnosing.

- In the questionnaire, we first explain its purpose and ask for careful completion, provide instructions on how to use the response scale for items, and add an example of filling it out.

- We start with easier and more attractive questions, place harder and less interesting ones in the middle, and end with usually simple factographic questions.

- We pay great attention to the formulation of each item and the use of individual words. We check how students understand the items and why they do not respond to some. We consult the questionnaire with colleagues. We revise unclear items.

- When constructing, we can consider that closed items are easier to evaluate (choice of answers) than semi-closed (choice of answers, plus the option of 'other – what') and open (write your answer).

- We do not use terms (several, usually, sometimes) that everyone interprets differently.

- We do not use double-barreled questions. It is then unclear what the student is responding to.

- We do not ask the student what they observe in the class (teacher is interested in the opinions of students) but their opinion regarding themselves (teacher is interested in my opinion). Data can then be aggregated for the class in the calculation of results.

- We use the vocabulary of the students.

- We only ask about what students can answer and what is meaningful to them.

- We make questions simple and clear, easy to answer.

- We do not use negative expressions (definitely not a double negative), as students often overlook or misunderstand them (e.g., this subject is not my favorite – yes/no). If a negative is necessary, we highlight it.

- We do not ask questions that evoke bias.

- In the questionnaire, we also use items where students respond on a scale (usually 1 – disagree/never to 5 – agree/always, N – do not know, cannot assess). These are scaled items. The number of points on the scale corresponds, for example, to the necessary variability of responses for the statistical analysis of data and the age of the students.

- We create questionnaires long enough to obtain the necessary information, but we try to keep them as short, simple, and concise as possible (usually a maximum of 30 minutes for adults). The possible length depends on the age of the students, their abilities, the topic of the questionnaire, etc.

Most questionnaires used by laypeople have basic errors in construction. Therefore, if you, as teachers, use questionnaires in your diagnostic or evaluative work, etc., it is recommended to study more information about their construction than what is presented here in this basic overview. Similarly, some questionnaires created by various experts may not necessarily be correctly constructed, and as teachers, you should be able to critically assess the construction of the questionnaire and use quality ones in your practice.

Processing Data from the Questionnaire

If we have a questionnaire for the entire class, we save (in the case of an online questionnaire) or transcribe the data into a data matrix, for example, in Excel (before importing it into a data analysis program). Each row corresponds to all the information about one student. In the columns, we have the questions (variables), and the individual responses to each question are coded with numbers (somewhere we have a manual for coding). How a data matrix looks and how to process data were covered last semester in the course "Research in Educational Practice."

We always check transcribed data before calculations. We verify that no errors occurred while transcribing from the printed questionnaire. We also look to see if students responded in an unusual way (e.g., 1-2-3-4-5-5-4-3-2-1, all 3s). In the questionnaire, we may have a so-called lie question, where everyone would actually answer yes (e.g., I like holidays), and if a student answers differently, it might indicate the unreliability of their responses. However, the problem of filling out is better addressed through good teacher-student relationships and students' trust in the meaningfulness of the questionnaire.

Questionnaires are usually processed statistically. Basic descriptive statistics (response frequencies, medians/averages, correlations, etc.) can be quickly calculated in a simple statistical program like SPSS, PSPP, Statistica, JASP. Very simple questionnaires for quick diagnosis of a student and class can be processed by hand (or some parts even in Excel, which, as a universal tool, is not easy for this type of data manipulation and descriptive statistics).

We calculate the frequencies of responses to individual questions. If the data on response scales (e.g., 1 – disagree to 5 – agree) are normally distributed (Gaussian curve), we can also calculate averages. More details can be found in the course "Research in Educational Practice" (SZ 6006).

 Interpretation of Results from the Questionnaire

A disadvantage of questionnaires is that we often do not know how to deeply explain the results. However, when diagnosing, we can ask the student why they responded in a certain way, as we continue to work with them as teachers. Questionnaires do not reveal what a student does, but what they report they do, which can differ from what they actually do (they may not be aware, estimates of frequencies of activities may be inaccurate, etc.). Also, if questionnaires are not filled out anonymously, responses may be biased. Students might also provide socially desirable answers that do not correspond to the truth, especially if a teacher selects the questionnaire for necessary planned informed measures in teaching. If you give a questionnaire to students, also show them that you respond to the results. Generally, it is advisable to work in an atmosphere of mutual trust.