Have you been asked to conduct a survey to complete a Research Methodology class assignment? Or are you currently working at a company and asked to conduct a survey to evaluate the quality of a new product? In fact, surveys have many very significant benefits, as long as they are made with a clear and transparent methodology. To make a quality survey, first try to determine the purpose of the survey and your target respondent/audience. After that, conduct the survey by contacting respondents via email, cell phone, post, or even meeting them in person. Finally, analyze your data and compile a final report on the survey results that were collected.
Step
Method 1 of 3: Creating Survey Questions
Step 1. Think about your goals
Before you start distributing surveys, first understand what is behind your survey. Are surveys being conducted to complete your assignment grades? Are surveys being conducted to get feedback on certain products? Once you know that, start thinking about the right target respondent or target audience and what questions you can ask to achieve the bigger goal.
- For example, the goal of your research is to determine how many students in your class will attend the school dance. Generally, these kinds of surveys only need to be answered with a yes or no, unless you want to know additional information such as motivation, clothes to wear, or other relevant factors.
- Make sure each question listed in the survey sheet can help you to achieve that goal. If necessary, revise your goals while designing the survey.
Step 2. Define survey parameters
Consider whether the respondent's identity is kept confidential, and whether the audience can see each survey result; specify when you will start and end the survey process; also determine how many people will be involved in the interview process or data analysis. The answer is easy if it's a personal project! Also specify the various instructions that you will include in the survey to guide the respondents.
- In fact, people can give more honest answers if their identities are kept secret. However, if that's the case, you won't be able to ask follow-up questions if needed.
- In the instructions, state how much time the respondent has to complete the respondent, and/or the tools they must use to fill out the survey (eg pencil only). Think of other options you could include.
- If desired, include a brief statement of the purpose of your research. This is very important if the survey is not conducted in person. By doing so, respondents will trust you more and be willing to respond to your survey.
Step 3. Tailor the question to your purpose
This is one of the most critical stages of compiling a list of questions. After determining the purpose of creating the survey, try to think about what information you need to collect to improve the survey? Have you had enough with basic and simple answers? Or do you need more elaborate answers?
If you want to know how someone is feeling, it's a good idea to ask them to give an open narrative response. However, if you want to analyze these emotions quantitatively, we recommend using ranking questions. For example, “How annoyed are you with X? Choose from 1 to 10 (10 means the most angry).”
Step 4. Understand the difference between open and closed questions
Think about it: Do respondents have the authority to give long answers? Or are they only allowed to choose one answer from the choices you provide? Once you've made your decision, start crafting survey questions and sorting out some of the best and most relevant questions.
An example of an open-ended question is, “Tell me about your childhood.” Meanwhile, an example of a closed question is, “Was your childhood happy? Answer with yes or no.” Closed questions will also limit the space allocated to answer respondents' questions
Step 5. Make sure you include demographic questions
If you want to analyze a respondent's final response while considering the demographic category, make sure you also create questions that are relevant to the demographic condition of the respondent. However, that doesn't mean you have to structure your questions based on each category; simply select the category most relevant to your goals.
For example, ask questions about the respondent's income, marital status, gender, ethnicity, age, or race. Generally, demographic questions of this kind will be arranged in the form of a list of choices that the respondent can circle or tick. For example, “Circle your marital status: Single or Married.”
Step 6. Pay attention to the order of the questions
It's best to direct respondents from easy questions to more difficult questions. Build respondent comfort before asking them to provide more private and intimate information.
We recommend placing demographic questions at the very beginning or end of the survey sheet. But in the worst case, if it is placed at the end and not asked directly, the respondent will choose to skip the question
Step 7. If you are working in a group, invite everyone to contribute
Try dividing up the tasks to make the questions fair. For example, ask each group member to think of several questions, combine all the questions that have been collected, and sort them to choose the most relevant questions. If everyone is focused on achieving the same goal, the list of survey questions will likely become more precise and focused.
Step 8. Keep your survey short
Ideally, the survey process should take 5-10 minutes. In other words, 5-10 minutes is the time required by the respondent to complete the survey. If the process takes too long, you will notice that the response rate of respondents will decrease drastically. However, if the survey process is forced to exceed 10 minutes, you can at least offer a reward to respondents so that they are still willing to complete the survey.
Step 9. Keep the survey safe
A qualified researcher must have a good safety record. Therefore, make sure you document all the details regarding the methodology used, the interview process carried out, and the final results obtained. In other words, everything that can be recorded and/or recorded must be documented! This process should start when you think about the purpose of the survey, and end when the final survey results are obtained.
If you are working with a research team, make sure you know who is responsible for each interview, what day the interview is being conducted, and other details. Also make sure you document each question that was deleted and why it was deleted
Method 2 of 3: Conducting a Standard Survey
Step 1. Create an incentive system
Trust me, quality responses will be easier to get if you give gifts or similar awards to respondents who are willing to fill out the survey. Therefore, consider holding a lucky draw, giving your thanks in public, giving away a promotional product or something more useful like a gift coupon.
Step 2. Conduct a pilot survey
Before sending survey sheets to respondents, try conducting a pilot survey on a smaller scale. Take advantage of friends, relatives, and those closest to you to be your test respondents. Have them fill out your survey sheet and provide feedback on the questions you asked, the survey filling process, etc. If possible, refine your survey based on their responses before sending it to actual respondents.
Also take the opportunity to observe the data and responses you receive. Was the result what you wanted? Do their responses address the main problem or question in your survey?
Step 3. Invite the respondent to meet in person
This is one of the best ways to conduct a survey, especially since respondents will generally be more responsive and the quality of the responses will be better. To conduct a face-to-face interview, you can first compile a list of contacts to be interviewed or simply select random respondents from the people you meet on the street. Do the process you choose until you reach the desired number of respondents.
- If the survey only needs to be done in class quickly, try walking around the class with a piece of paper. After that, ask your questions to the respondents and record their answers using a tally mark.
- Understand that in-person interviews are often seen as more personal. As a result, the interview situation will often feel awkward, especially if the questions asked are sensitive. It is feared that this awkwardness can reduce the accuracy of your respondents' answers.
Step 4. Use an online survey program if you wish
The online survey program is one of the newest methodologies for conducting surveys. If you use this method, respondents will be immediately led to enter one of the many survey sites available, such as Google Surveys, Survey Monkey, Dot Survey, and Key Survey. On the site, there are details that they need to fill in and complete.
- Most sites even provide basic templates that you can use for free. However, you may be asked to pay if you want to use the same survey template again, complete survey details, or increase the range of respondents. However, this is actually a relatively cheap option for you.
- These survey sites can even help you to analyze the data that has been collected.
Step 5. Calculate the result
Once the survey results are complete, take a look at the data you have collected and decide how to report it. For example, you can report survey data in the form of graphs, tables, or charts containing detailed statistics. If the survey is for business purposes, you will most likely be asked to present a formal report to your boss.
Method 3 of 3: Conduct a Scientific Survey
Step 1. Determine the sample size to be used
In other words, determine the number of respondents you need and how to avoid biasing the survey results. Generally, it is easier if you have random participants or select respondents based on the demographic conditions of the target audience.
- For example, Pew Research limits the number of international respondents to 1,000 people per country. Although it sounds small, in fact this limit actually helps them to reach more countries.
- Determine the number of respondents realistically. Consider the resources and time you have available to conduct the survey. Remember, data quality doesn't go hand in hand with the number of surveys!
Step 2. If necessary, obtain approval from the authorized ethics committee
If you operate under the auspices of a university or company, you will most likely need to obtain approval to conduct research from an authorized ethics committee. Generally, this is necessary because research surveys involve interactions with humans. When your request is reviewed by the ethics committee, make sure you provide as much information as possible regarding the purpose of the survey and the methodology used.
Step 3. Get sponsors
Remember, one of the factors you should take into account is the cost of the survey, if your target audience is wider than just one classroom. If the survey is intended for scientific research, try submitting a funding request to the government, university, or other regulatory authority. You can even target specialized organizations that are engaged in your research field. In general, you will need 400 thousand rupiah for each respondent to a survey conducted by telephone.
Step 4. Contact respondents via email for a faster survey option
In fact, it is one of the most preferred methods of communication today. That is why, many people choose to use it to conduct surveys. Besides being easy and cheap, the process of sending email is also very fast, although sometimes you need to buy an email contact list. Using this method, you can generally reach your targeted target audience, and have them return the completed survey sheet to your email address or link them to a special link. At worst, your email will be deleted by a respondent who refuses to respond.
Step 5. Contact the respondent by mail if you prefer to use the traditional method
Although somewhat ancient, sending the questionnaire directly to the respondent's address can still be applied. Apart from making it easier for you to reach a wider geographic area, this method also feels more friendly to older respondents who are not used to using email and similar technologies. However, this method also has its drawbacks, namely that you need to incur substantial shipping costs and wait longer to collect responses.
Step 6. Call the respondent by telephone if you have the telephone number
If you want to conduct a survey by telephone, try to think about whether it is better to contact respondents via cellphone or landline? Also look for ways to get the respondent's phone number (for example you can buy the respondent's phone list). Conducting telephone surveys is one of the cheapest methods, but has the highest rejection rate because people generally feel less comfortable being contacted via personal channels such as the telephone.
Step 7. If you have large enough funds, work with a research firm to conduct the survey
Browse the internet to find the closest research firm in your city. Although it really depends on your budget, it's a good idea to work with a professional group to interview respondents, or even create survey questions at the same time. This option is worth checking out if you need expert help to create quality surveys and streamline the process.
Read all the policies applied by the company you work for. Make sure the privacy handling information is also included in the policy. It is also advisable to arrange a confidential agreement to ensure that the entire process and the final survey results are well protected
Step 8. Monitor your assigned interviewers
Working in the field is challenging. That's why, the people you assign must be individuals who have been professionally trained to conduct surveys in order to get the best results. Make sure you ask for their personal contact, especially to monitor the situation on the ground.
Beware, some researchers may need additional training in counseling if they need to ask respondents personal and emotional questions
Step 9. Follow the rules that apply in your country
Make sure your research is done under applicable laws before publishing it! This is especially mandatory if you need to contact the respondent without asking for their consent first. Generally, there are age, time, and communication methods that you need to understand before conducting a survey.
For example, some laws prohibit researchers from using the automatic dialing feature to call someone
Step 10. Use survey results for professional use
Most likely, the final results of your survey will be used for scientific development purposes. Some fields of study, such as sociology, provide the ground for journal publications, presentations at conferences, and seminars. Whatever your survey approach, try to find ways to share your findings and analysis with a wider academic (and even general) audience.
Tips
- The more respondents, the better your survey results will be. In other words, the results you get will definitely be better and more comprehensive if your respondents are 100 people instead of just 10 people.
- Be patient in compiling and completing the survey. Most likely, you will encounter survey sheets that are not filled out or similar problems when evaluating survey results.
Warning
- As you formulate questions and train interviewers, make sure you also pay attention to issues related to language or translation.
- Make sure the interviewer you assign doesn't ask more than one question at a time. Be careful, having to answer many questions at the same time can reduce the accuracy of respondents' answers.
- Always remember that any information you request may be personal to the respondent. Therefore, try to establish a specific policy regarding respondent confidentiality, and teach each interviewer you assign to each interviewer how to respond to respondent concerns regarding the security of their identity and privacy.
