STEP BY STEP INSTRUCTIONS TO DESIGN AND DEVELOP A QUESTIONNAIRE FOR A FINAL YEAR PROJECT
A questionnaire is a strategy for gathering information whereby a respondent gives answers to a progression of questions given to him or her by a researcher or a research student. And as straightforward or simple as it might appear or sound, most students still think that it's very hard to frame or design a decent research questionnaire for their research project works.
Although, each educational field of study has its own format for the design of its research questionnaires, at times this might not be an order from the school but the choice of the supervisor.
Before a research student will be able to develop a decent questionnaire properly he or she must first of all be able to understand his or her final year project topic or project topics properly, before deciding on the kind of questions he or she will like to ask his or her respondents.
A questionnaire is a strategy for gathering information whereby a respondent gives answers to a progression of questions and to build up a decent questionnaire that will gather the information you need takes effort and time. However, by adopting a step by step strategy to questionnaire development, you can think of a successful way to gather the information that will address your remarkable research question.
Questionnaire design is a multistage process that requires attention to many details at once. Designing the questionnaire is a bit complicated because the studies can ask about topics in varying degrees of detail, questions can be asked in different ways, and questions asked earlier in research may influence how people respond to later questions. Researchers also are often interested in measuring change over time and therefore must be attentive to how opinions or behaviors have been measured in prior studies.
Maybe the most important part of the research process is the development or creation of questions that accurately measure the opinions, experiences, and behaviors of the public. Accurate random sampling and high response rates will be wasted if the information gathered is built on a shaky foundation of ambiguous or biased questions. Creating good measures involves both writing good questions and organizing them to form the questionnaire.
The diagram below shows a clear and decent research process to follow when designing a decent questionnaire
The model of the research process
Experiences and motivation: You may be given a title for a project or you may choose your own. Your own background and interests will help you determine the research question.
Literature review: It is very important to know what research has been done in your field so that you can ascertain the value of your contribution (your position in a conceptual framework). Knowledge of any gaps in the literature may guide you to the best research question for your study.
Strategies: Choose a strategy or strategies that will suit your research.
Data generation methods: You may be working from secondary data but most likely will need to gather data by one of these methods.Data analysis: Use qualitative or quantitative methods depending on the type of data collected.
Designing and administering questionnaires
Many students do projects that use a questionnaire survey to collect data. These must be carefully designed so that the data you collect is meaningful and can be analyzed.
This stage has to follow a careful plan of what your research question is and what hypothesis /hypotheses are you testing.
The control group in research
You may want to include a control group.
E.g. if you wanted to know the extent to which disabled children from a special school use computers you could survey a group of children of the same age from ordinary school as a control group.
Types of data
- Surveys can collect written data (qualitative) and measurement data or preferences on a sliding scale (e.g.using the Likert scale).
- These latter two categories of data are classed as quantitative data.
- Can be used in conjunction with interviews.
- The interview can produce qualitative data.
- The questionnaire can produce quantitative data.
The nature of qualitative research
- Qualitative research uses non-probability sampling as it does not aim to produce a statistically significant result.
- Qualitative research aims for breadth as well as depth.
Purposive sampling is often used in qualitative research. When using purposive non-random sampling the number interviewed is less important than the criteria used to select them.
Determine what you need to know
- What exactly do you need to find out?
- What type of question do you want to ask?
- The more structured the question the easier it is to analyze.
- Types of questions (from Bell (1993) p 76, after Youngman (1986))
Linking the survey questions to theory
If you decide to collect survey data via a questionnaire for your project then you need to be able to design the questionnaire with sound theoretical underpinning, a description, and justification of your choice of questions, a sampling strategy and descriptions of your participants, and the number surveyed. And additionally, you must also describe the procedure for collecting the data and consent and ethical considerations in dealing with your participants.
- What exactly do you need to find out?
- What is your research question? This is a high-level question.
- What are the hypotheses arising from the research question? You will collect data to answer these questions.
- What type of question do you want to ask? The more precise and focused the question the easier it is to analyze.
Pros and cons of questionnaires
–Less cost and time
–Low cost to distribute especially online
–People like answering structured questions without much need for writing
–Social desirability in answering
–People may deliberately lie
–Select from a list you give them
–Offer selected categories such as age categories e.g. 16-25, 26-35 etc
–The respondent is asked to put something in rank order
–Scales are devices to elicit strength of feeling or attitude. A straightforward attitude scale is a Likert scale
Ambiguity, imprecision, and assumption
- Words that appear common sense to you may have a totally different meaning to others
- Be precise in your wording
- E.g. you ask “What type of school does your child attend? And the respondent chooses from a list of types of school.
- You have assumed that the mother has one child-but she may have several and at different levels of schooling (primary and secondary)
Also, read: HOW TO WRITE EFFECTIVE RESEARCH PROJECT ABSTRACT
Developing a questionnaire
Identify the goal of your questionnaire. What kind of information do you want to gather with your questionnaire? What is your main objective? Is a questionnaire the best way to go about collecting this information?
- Come up with a research question. It can be one question or several, but this should be the focal point of your questionnaire.
- Develop one or several hypotheses that you want to test. The questions that you include in your questionnaire should be aimed at systematically testing these hypotheses.
Choose your question type or types. Depending on the information you wish to gather, there are several possible types of questions to include on your questionnaire, each with unique pros and cons. Here are the types of commonly used questions on a questionnaire:
- Dichotomous question: this is a question that will generally be a “yes/no” question, but may also be an “agree/disagree” question. It is the quickest and simplest question to analyze but is not a highly sensitive measure.
- Open-ended questions: these questions allow the respondent to respond in their own words. They can be useful for gaining insight into the feelings of the respondent but can be a challenge when it comes to analysis of data. It is recommended to use open-ended questions to address the issue of “why.”
- Multiple choice questions: these questions consist of three or more mutually-exclusive categories and ask for a single answer or several answers. Multiple choice questions allow for easy analysis of results, but may not give the respondent the answer they want.
- Rank-order (or ordinal) scale questions: this type of question asks your respondent to rank items or choose items in a particular order from a set. For example, it might ask your respondents to order five things from least to most important. These types of questions force discrimination among alternatives but do not address the issue of why the respondent made these discriminations.
- Rating scale questions: these questions allow the respondent to assess a particular issue based on a given dimension. You can provide a scale that gives an equal number of positive and negative choices, for example, ranging from “strongly agree” to “strongly disagree.” These questions are very flexible, but also do not answer the question “why.”
Develop questions for your questionnaire. The questions that you develop for your questionnaire should be clear, concise, and direct. This will ensure that you get the best possible answers from your respondents.
- Write questions that are succinct and simple. You should not be writing complex statements or using technical jargon, as it will only confuse your respondents and lead to incorrect responses.
- Ask only one question at a time. This will help avoid confusion
- Beware of asking for private or “sensitive” information. This can be something as simple as age or weight, or something as complex as past sexual history.
- Asking questions such as these usually require you to anonymize or encrypt the demographic data you collect.
- Determine if you will include an answer such as “I don’t know” or “Not applicable to me.” While these can give your respondents a way of not answering certain questions, providing these options can also lead to missing data, which can be problematic during data analysis.
- Put the most important questions at the beginning of your questionnaire. This can help you gather important data even if you sense that your respondents may be becoming distracted by the end of the questionnaire.
Restrict the length of your questionnaire. Keep your questionnaire as short as possible. More people will be likely to answer a shorter questionnaire, so make sure you keep it as concise as possible while still collecting the necessary information. If you can make a questionnaire that only requires 5 questions, does it.
- Only include questions that are directly useful to your research question. A questionnaire is not an opportunity to collect all kinds of information about your respondents.
- Avoid asking redundant questions. This will frustrate those who are taking your questionnaire.
Identify your target demographic. Is there a certain group of people who you want to target with your questionnaire? If so, it is best to determine this before you begin to distribute your questionnaire.
- Consider if you want your questionnaire to collect information from both men and women. Some studies will only survey one sex.
- Determine whether you want your survey to collect information from both children and adults. Many surveys only target certain age ranges for which the questions are applicable.
- Consider including a range of ages in your target demographic. For example, you can consider a young adult to be 18-29 years old, adults to be 30-54 years old, and mature adults to be 55+. Providing an age range will help you get more respondents than limiting yourself to a specific age.
- Consider what else would make a person a target for your questionnaire. Do they need to drive a car? Do they need to have health insurance? Do they need to have a child under 3? Make sure you are very clear about this before you distribute your questionnaire.
Ensure you can protect privacy. Make your plan to protect respondents’ privacy before you begin writing your survey. This is a very important part of many research projects.
- Consider an anonymous questionnaire. You may not want to ask for names on your questionnaire. This is one step you can take to prevent privacy, however, it is often possible to figure out a respondent’s identity using other demographic information (such as age, physical features, or zipcode).
- Consider de-identifying the identity of your respondents. Give each questionnaire (and thus, each respondent) a unique number or word, and only refer to them using that new identifier. Shred any personal information that can be used to determine identity.
- Remember that you do not need to collect much demographic information to be able to identify someone. People may be wary to provide this information, so you may get more respondents by asking less demographic questions (if it is possible for your questionnaire).
- Make sure you destroy all identifying information after your study is complete.
- Avoid any ambiguity in meaning
- Be precise when asking questions-think of the range of possible answers and be sure that your question will elicit the full range
- You bring to your questionnaire design your cultural values e.g. you cannot assume that everyone will be free on a Saturday to answer a survey –some religions worship on a Saturday.
- Memory can play tricks.
- If you ask about something that happened a long time ago like games played at school a younger person would probably remember more than someone older.
- It depends on why you want this question answered-if it is critical that you get as much data from the person as possible then perhaps provide a list of games they can choose from.
- Do not ask leading questions.
- These are usually subjective and often emotive and are guiding the respondent in one direction.
- Double questions should never be asked
- Do you like apples and oranges?
- Yes to one? Or both?
- How do you analyze data from this question?
Appearance, layout and length
- Questionnaires should be typed or printed.
- Instructions should be clear.
- Spacing between questions helps the reader.
- If you want a limited number of sheets –reduce the copy.
- Keep all response boxes in line towards the right of the sheet.
- Start with straightforward, easy-to-complete questions.
- Be critical of your questionnaire.