Survey Design Flashcards
(25 cards)
What is a survey design?
“It’s a systematic method for gathering information that reflects on a populations attitudes, behaviors, opinions, and beliefs that cannot be observed directly (constructing quantitative descriptors of some attitudes). A survey is successful when people’s answers are as close to their thoughts and beliefs as possible. “
How can survey design be considered a research methodology?
Because the research is governed by a sequence of steps defining; determining the info; planning; designing; addressing; coding; statistical analysis; presentation of results
How can survey design be considered a data collection technique?
Because it has a set of questions known as a questionnaire, whose development follows a series of rules and principles.
A survey methodology is non-experimental method…
whose characteristic is that information is collected in the absence of manipulation or intervention by the researcher, using consistent and standarized measurement procedures for all subjects. It can be quantative or qualitative data
What are the objectives of a survey?
1) to understand the characteristics of a population and the relationships between them. 2) ability to generalize results to a population based on data obtained from a representative sample drawn from that poputlation 3) no need to assignment in terms of ethical reasons. BENEFITS from obtaining information we cannot see or maybe observe like opinions, beliefs, or attitudes.
Types of survey designs
1) Population (Census and Sample) 2) Survey (Descriptive and Analytic) 3) Time Dimension (Cross-sectional and Longitudinal)
Types: Population Census
Data from every individual or unit in a populaiton (indicated when the population is relatively small and every element can be located)
Types: Population Sample
Subset of individuals, items, or units selected from a larger population
Types: Surveys that are descriptive
Aimed at obtaining specific information about a population, or subgroups within it, in terms of descriptive statistics
Types: Analytical that are descriptive
Aimed at studying the relationships between phenomena in the real world, analysing the relationships between different charactersitics and variables that affect a group of a population to varying degrees
Describe the planning the research phases of survey?
Planning the research: Desiging the research instruments (quesitonnairs) here we examine what we are trying to achieve with the survey results. Purposes should aim towards describing the characteristics of the population or their specific context (We decide the types of quesitons, wording, and sequencing. We also decide how to administer Traditional or online.)
What is survey sampling methods?
a small group of people or units are chosen from a bigger population so that a survey can be done. There are probability sampling and non-probability sampling.
What is probability sampling?
It is a sampling that involves random selection allowing to make strong statistical inferences about the whole group. We have simple random sampling: way of selecting participants that has the same chance of being selected as any other member. We also have clusted samping: when the pop. is divided in clusters, and then clustes are randomly selected for inclusion in the sample (might be pre-existing like zip codes or students) used to gather infromation from a dificult situation
What are non-probability sampling methods?
It is a sampling that uses a non-random criteria like the availability, geographical proximity, or expert knowledge of the individuals you want to research in order to answer a RQ. Their used when the population parameters are either unknown or not possible to individually identify. (Hight risk for research biases that prob. sampling)
What are the types of non-probability sampling?
Quota sampling is a non-random sampling method where researchers aim to include a specific number (or proportion) of people from different subgroups in the population. Units share specific characteristics. Snowball Sampling: People recruited to be part of a sample are asked to invite those they know to take part, which are then asked to invite their friends and family. (downside is representativeness)
What is the questionnaire construction?
The questionnairte indirectly measures what people know, think or judge abour something, and this not a simple task
Within the survey we have to think about the…
1) Types of questions (closed or open-ened questions). 2) the content of the questions. 3) the phrasing of the questions. 4) the ordering and layout of the survey. 5) distribute the survey and collect responsed. 6) analyze the survey results.
Types of survey questions
- Open-ended questions
Allow respondents to answer freely in their own words.
Example: “What do you think about public transportation?” - Closed-ended questions
Provide predefined answer choices.
Includes yes/no, multiple choice, or rating scales.
Example: “Do you own a car? (Yes/No)” - Semi-open or semi-closed questions
Combine fixed answer options with space for additional comments.
Example: “What is your preferred mode of transport? (Car, Bus, Bike)” - Demographic questions
Gather information about personal characteristics like age, gender, education, or income.
Example: “What is your age?” - Attitudinal questions
Assess opinions, beliefs, or feelings about a topic.
Example: “How strongly do you agree with the statement: ‘Public transport is reliable’?” - Behavioral questions
Ask about past actions, current habits, or future intentions.
Example: “How often do you use public transportation per week?” - Introduction questions
Used to ease the respondent into the survey topic.
Example: “How familiar are you with our local bus system?” - Filter questions
Guide respondents to relevant follow-up questions based on their answers.
Example: “Do you have a driver’s license?” → If yes, ask about driving habits. - Control questions
Help check for consistency and data quality or control for bias.
Example: Asking the same question in different ways to verify responses.
Before the real thing we should conduct a…
Pilot test on small and non probabilistic samples of participants to observe issues related to its implementation. Helps understand difficulties participants will encounter with the instrument, results can also warn us of some response biases.
What are the different ways we can collect interview data?
Face-to-face, telephone, Postal Mail, Email, or Social Media
How do we process data?
We go through:
1) Data cleaning (errors, inconsistencies, or missing values)
2) Data entry (if collected manually)
3) Data coding (converting qualitative data into numerical values)
4) Data Analysis (descriptive statistics to summarize the data, inferential statistics to test hypothesis, or advanced techniques regession analysis)
5) Interpretations (putting RQ and the findings together to draw conclusions) 6) reporting
What are errors in surveys?
They can arise from various stages in the research process and can impact the validity of the data collected. Commonly in sampling, measurement instruments, data processing.
Sampling errors happen when…
happen when the sample is not done properly or when the estimation of the sample size is incorrect
Errors in the design of the measurement instrument happen when…
These errors can be associated with the questionnaire itself if its poorly formulated or even incorrectly administrated
