Assignment 4: Methodology Section
Slide 1: Intro to Surveys
Surveys are one method of collecting data for research that includes participants responding to written questions and prompts. Survey research includes questionnaires, or instruments containing the questions and measurements in a self-administered format, the respondent, or person taking the survey, and a knowledge of the response rate, or the percentage of persons who completed the survey in your sample.
Because surveys are self-administered, researchers must pay close attention to the way that questions are developed and whether or not they are understandable to the respondents who will be surveyed. Questions on a survey can be either closed or open-ended. Being open-ended means that a respondent is able to fill in their own personalized response to a question. Often, this requires the researcher to ask for specific information or for a narrative response. Alternatively, close-ended questions are when respondents choose from a list of possible responses that were provided by the researcher. Close-ended questions are also called fixed-choice or forced-choice questions.
Both open-and close-ended questions have advantages and disadvantages. Close-ended questions allow for a quick response, consistency, and are easier to statistically analyze, but may cause results to be biased if a participant is forced to choose an option that does not fit with their actual opinion or perspective. Open-ended questions allow for detailed responses, and reduce biases related to preconceived answers, but must be thoroughly reviewed before they can be used for analysis.
Slide 2: Principles for writing questions
In order to make your survey as high-quality as possible, you should always put effort into writing clear and meaningful questions, avoiding confusing phrasing, minimizing bias, avoiding making disagreement, and maximizing response categories while minimizing opportunities for participants to float.
Essentially, wringing clear and meaningful questions means that surveys should be generalizable and understandable to many people. You cannot rephrase a survey question that someone does not understand because it would bias the results, so always strive to make sure that your questions are easily interpretable and your meaning clear. In order to increase this clarity, make sure to use correct grammar and keep your phrases shorter. If your survey includes complex ideas, separate them into different questions and provide definitions when necessary, and always avoid vagueness. Being more specific will reduce confusion and bias.
You must also strive to avoid double negatives, double-barreled questions, and your answers should always be mutually exclusive and exhaustive. Having double negatives in your questions will increase the likelihood that a respondent will misinterpret your question. While professors may use double negatives on tests to trip-up students, doing so in surveys will negatively affect your results. An example of a question that used double negatives is: ‘Do you disagree that there should not be a death penalty?’ Instead of making this phrasing complicated, you should instead ask, ‘Do you agree with having a death penalty?’
In order for your survey questions to be mutually exclusive and exhaustive, you should always make sure there are no overlaps in your response categories, and that they cover all options that a respondent could pick. If you ask the question of how many times a person has been arrested and your options are 0-1 or 1-3, then they are not mutually exclusive, because the number one was included in both categories, and they are not exhaustive, as a person could have been arrested more than three times. Thus, a better choice of options would be: 0, 1-3, 4-6, 7 or more.
Slide 3: Types of Questions
Likert questions are often used with declarative statements rather than with questions. The options available are generally on a five point scale, ranging from strongly agree to strongly disagree, although there has been an increased movement to remove the neutral section. You also have the opportunity to include not applicable options if you feel that some statements will not be necessary for some respondents.
Filter questions and skip patterns are used when questions may only apply to some respondents. It allows the survey to become quicker for those with whom the measurements do not apply, and it reduces respondent burden. If your survey is not electronic with automatic skip patterns, make sure your survey is clearly labeled in terms of which questions should be skipped and to which question a respondent should move to if they are not required to answer a certain section.
Lastly, demographic questions are those that provide a basic description of your respondents. They include questions that address age, sex, gender, race/ethnicity, education, income, religion, employment status, occupation, region of residence, sexuality, ect… While you do not need to include every demographic variable in your research, you should include some in order to make sure your sample is as representative as possible.
Slide 4: Types of Surveys
Surveys can be either self-administered or administered by the researcher. Based on the population you are trying to reach and the number of participants you are attempting to recruit, your method of surveying may be different. Self-administered surveys are when respondents fill out their own questionnaires, and are generally the most economical way to survey large numbers of people. Electronic surveys are becoming increasingly used as technology advances, especially as surveying programs have developed methods of transferring results straight into statistical packages, however, it is more difficult to know that the electronic surveys are representative unless you have a list of emails or other forms of identification before hand from which you can draw a representative and random sample.
Researcher-administered surveys may increase the quality of the data a researcher gathers, however, they may take more time and resources to procure. Not only does the researcher have to set aside large amounts of time to complete these forms of surveying, but they can also be expensive as those administering the survey require training and may need to travel to different locations to administer the questionnaires.
In the end, the type of survey that you administer is going to depend on the population you are trying to reach, your sample size, the topic you are trying to address, and the time and resources you have available to complete the project.
Slide 5: Module Wrap-Up
After reading the texts and listening to the lecture prepared for this module, you should be confident in your ability to complete the learning objectives from the unit.
In particular, you should be able to elaborate on how a questionnaire or interview schedule is the foundation of surveys, and debate on various survey layouts and the implications of effective survey design. Consider writing up your own practice survey questions to get used to avoiding double negatives, double-barreled questions, and making sure you can develop questions that are mutually exclusive and exhaustive.
Your second discussion board assignment will be due at the end of this module. Remember, you must make your own response to the question, and complete two further responses to various classmates. Make sure to check blackboard or the syllabus for further guidelines regarding this assignment and do not hesitate to post questions in to the interactive discussion board for feedback from your classmates or professor.