Research Methods: Online Surveys and Interviews

Podcast

Play an AI-generated podcast conversation about this lesson

Questions and Answers

What is one method to reduce concerns about disguising research?

Including as many detailed questions as possible

Allowing respondents to remain anonymous

Letting respondents know the study is 'blind' and why (correct)

Avoiding any form of debriefing

Which of the following is a notable disadvantage of telephone interviewing?

Higher cost than personal interviews

Challenges related to mobile technology (correct)

Increased ability to connect with respondents

Improved response rates

Which of the following statements about personal interviews is false?

They tend to be time- and cost-intensive

They involve a direct, face-to-face conversation

They allow for great flexibility in questioning

They are typically low in interviewer bias (correct)

Which of the following is not a method for optimizing online surveys?

Using complex technical terms Signup and view all the answers

What is a key advantage of online surveys compared to traditional methods?

Typically quick and inexpensive Signup and view all the answers

Which of the following characteristics is true for online surveys?

They have become increasingly popular in recent years Signup and view all the answers

What is a method for increasing engagement in online surveys?

Keeping the survey interactive Signup and view all the answers

What is one key factor influencing response rates in online surveys?

The perceived incentives for completion Signup and view all the answers

What is one of the most important actions to take before launching a questionnaire?

Conduct a pretest Signup and view all the answers

Which of the following is a key consideration when creating closed-ended questions?

Include an option for 'don't know' if applicable Signup and view all the answers

Which of the following does not apply to sampling error?

Should be the focus as the single most important error Signup and view all the answers

Which wording practice should be avoided when formulating survey questions?

Employing ambiguous language Signup and view all the answers

Which of the following is a source of error besides sampling error?

Noncoverage Signup and view all the answers

What is a convenience sample?

A sample taken from respondents readily available at a location Signup and view all the answers

Why is sampling error considered less troublesome compared to other errors?

It can be estimated and accounted for Signup and view all the answers

What distinguishes a census from a sample?

A census examines all population members, while a sample examines a subset Signup and view all the answers

Why is probability sampling preferred?

It facilitates statistical inference about the population Signup and view all the answers

What is the primary goal in sample surveys regarding sources of error?

To decrease total or overall error Signup and view all the answers

The degree of nonresponse in a survey is an indicator of what?

Overall quality of the survey Signup and view all the answers

What is snowball sampling typically used for?

Accessing special, hard-to-reach populations Signup and view all the answers

Which option describes a potential drawback of double-barrel questions?

They can confuse respondents about what is being asked Signup and view all the answers

Which strategy is NOT effective in handling nonresponse?

Ignoring nonresponse cases completely Signup and view all the answers

Which of the following statements about nonresponse is true?

Nonresponse can be common even in large samples Signup and view all the answers

What is a common result of recording errors in a survey?

Inaccurate data analysis Signup and view all the answers

What is the first step in preparing for regression analysis?

Identify the Y and X variables Signup and view all the answers

Which statistical method is used to compare means between two categorical and one continuous variable?

Independent samples t-test Signup and view all the answers

In regression output, what does Multiple R represent?

The strength of the relationship between Y and X variables Signup and view all the answers

What is the role of the categorical variable in an independent samples t-test?

To classify data into two groups Signup and view all the answers

What must be done to the X and Y variables before performing regression analysis?

They must be adjacent and any missing values filled Signup and view all the answers

Which of the following statements is true regarding the use of pivot tables?

They can examine relationships between both categorical and continuous variables. Signup and view all the answers

What does Multiple Regression involve?

Two or more predictor variables Signup and view all the answers

Which of the following is not a characteristic of observational research?

It accurately predicts future behaviors. Signup and view all the answers

Which of these factors is most likely to improve response rates?

Survey length Signup and view all the answers

Which strategy is least effective in addressing respondent nonresponse?

Place sensitive questions right up front Signup and view all the answers

Which of the following tasks should be performed in the data preparation stage to ensure quality?

Checking for consistency across responses Signup and view all the answers

Which of the following is typically included in preliminary steps of data analysis?

Editing or cleaning the data Signup and view all the answers

What method is commonly used to determine whether respondents are not thinking carefully about their answers?

Identifying straight lining in their responses Signup and view all the answers

Which of these actions would best ensure that data quality standards are maintained?

Converting all responses to consistent formats Signup and view all the answers

Which of the following techniques can enhance a survey's ability to lower nonresponse rates?

Personalization of survey invitations Signup and view all the answers

What is critical to assess in the data preparation stage in order to maintain quality?

Deleting records with high nonresponse rates Signup and view all the answers

What does the coefficient of multiple determination (R2) indicate?

How much variation in the dependent variable is explained by the independent variables Signup and view all the answers

What type of variables does a Chi-Square Good-of-fit Test typically analyze?

Categorical variables Signup and view all the answers

Which statement correctly describes the difference between univariate and multivariate analysis?

Univariate analysis examines one variable at a time, while multivariate analysis looks at relationships among multiple variables. Signup and view all the answers

What is the purpose of a one-sample t-test?

To compare the mean of a sample to a known value Signup and view all the answers

Which of the following is NOT one of the three MUST HAVES for effective communication?

Using complex jargon for clarity Signup and view all the answers

What is meant by research reports serving as an archive?

They are records of completed research for reference and future use. Signup and view all the answers

In regression analysis, which statistical measure is used along with the p-value to determine model significance?

F statistic Signup and view all the answers

What is a fundamental rule in research presentations aside from knowing your audience?

Thorough knowledge of the topic Signup and view all the answers

Study Notes

CS 201 Exam 2 Review

Exam 2 review focuses on data analysis.

Data Analysis Review

Data analysis should begin by asking questions about the data.
Questions need to be answered by considering variables, the number of variables, and type of variables.
Examples of analysis considerations include:
- How many members used the circuit?
- What is the average monthly revenue?
- Describe members' visits/pool usage.
- How often do pool users visit AFC?
- Is member fitness or social aspects more important?
- Relationship between likelihood to recommend and visits?
- Impact of age, likelihood to recommend, and education on fees.

Types of Analysis

Univariate Analysis: Focuses on a single variable, describing the data.
- Examples: How many members used the exercise circuit, average monthly revenue of AFC?
Bivariate Analysis: Examines the relationship between two variables.
- Examples: Pool users versus non-pool users visiting AFC, social aspects or fitness importance rating.
Multivariate Analysis: Examines the relationship among three or more variables.
- Examples: Influence of age, likelihood to recommend, and education on fees.

Excel Commands

Count of members using exercise circuit: Use the =countif function.
Average monthly revenue: Use the average function.
Describe member visits: Run Descriptive Statistics Summary in Excel.
Compare pool users versus non-users: Run an independent samples t-test.
Compare social and fitness importance: Run a paired samples t-test.
Relationship between likelihood to recommend and visits: Use the =correl function for correlation.
Influence of age, likelihood to recommend, and education on fees: Use regression analysis.

Observation Research

Observation research is less versatile than other methods, observing behavior in the moment and not revealing "why." It cannot measure future intentions. It is highly accurate, but subjective.
It might require waiting for behaviors to occur.
Direct observation: Observing actual activity.
Indirect observation: Observing the outcome of activity.

Primary Research

Degree of structure, disguise, setting, and method of administration are factors to consider when designing primary research.
Questions should be standardized so responses and questions are consistent, with 'fixed alternative' being used instead of 'open-ended', wherever possible.
Highly structured questions are best when possible responses are known and limited in number.

Considerations for Developing Questionnaire Questions

Always ensure that questions are necessary and not too revealing.
If more than one question is needed to measure one piece of information, use multiple questions instead.
Make sure respondents have the necessary information to answer questions.
Consider whether respondents will be willing or able to provide the information.

Filter Questions

Filter questions are questions that determine if a respondent has the relevant knowledge or qualifies for the study.
They are used to screen respondents.

Questionnaire Pretesting

Pretesting your questionnaire before launching it is a crucial step.

Closed-Ended Questions

Include "don't know" options is important for closed-ended questions.
Responses must be exhaustive (including "other" options).
Responses need to be mutually exclusive to prevent confusion.
Be aware of response order bias.

Wording of Questions

Avoid simple words when possible.
Avoid ambiguous wording and phrasing.
Avoid leading questions that suggest an answer.
Avoid assuming respondent knowledge.
Avoid double-barrel questions that ask multiple things. (e.g., asking about multiple things simultaneously in a single question.

Nonprobability Sampling

Convenience, Judgment, and Quota sampling are types of non-probability sampling.

Census vs. Sample

Census: Collects data from every member of a population.
Sample: Collects data from a subset of a population.

Probability Sampling

Probability sampling is preferred because sampling error can be estimated.

Snowball Sampling

Snowball sampling is a judgment sampling technique to find special or hard-to-find populations.

Precision vs. Confidence

Precision measures the degree of error in an estimate.
Confidence measures the degree to which an estimate approximates the true value.

Sampling Considerations

Precision and variability affect sample size requirements, with a greater degree of precision and more variability requiring a larger sample size.
The size of the population does NOT affect sample size.

Sampling Frame

A sampling frame is a list of elements from which a sample is drawn.
Example: customer database, member directories.

Approaches to drawing a Sample

Probability sample: Every population element has a known chance of being selected. Examples include Simple Random, Systematic, Stratified, and Cluster sampling (including area).
Nonprobability sample: Relies on personal judgment. Examples include Convenience, Judgment, and Quota sampling

Target Population

The target population comprises all individuals or entities meeting specified criteria for a study.

Improving Precision

Increase the sample size.
Decrease the confidence level (e.g., from 95% to 90%).

Random Error

A temporary, personal, or measurement situation error.
Affects a measurement in irregular ways.
Difficult to control.

Data Preparation Stage

Convert all data to consistent units (e.g., months to years, dollars to cents).
Evaluate the degree of non-response and remove records with more than 50% missing data.
Check across responses for consistency.
Look for straight-lining (all responses being the same).
Check if branching questions were correctly followed.

Preliminary Data Analysis

Edit or clean the data.
Code and convert data into consistent symbols.
Execute frequency analysis.
Identify outliers.

Types of Primary Research

Descriptive: Describes a group or the relationship between variables.
Exploratory: Gains ideas or insights to define the problem or opportunity for study.
Causal: Determines cause-and-effect relationships.

Data Analysis Considerations

Determine the proper type of analysis (e.g., univariate, bivariate, multivariate).
Identify appropriate measurement levels (nominal, ordinal, interval, ratio).

Systematic Error

An error in measurement that affects it consistently. Also called constant error.

Chi-Square Good-of-Fit Test

A statistical test used with categorical variables to compare observed frequencies to expected frequencies.

Independent Samples t-test

A statistical test used to determine if there is a difference between the means of two groups when one variable is categorical and the other is continuous.

Correlation

Measures the degree of linear association between two continuous variables.
Correlation is not causation.

Regression

Identifies a "best fit" between predictor and outcome variables.
Shows the individual impact of predictor variables.
Provides a model that shows the combined impact of predictor variables (coefficient of multiple determination (R2)).
Uses an F-statistic and p-value to assess the model's significance.

P-value

A p-value is the probability of obtaining a given result if the null hypothesis was true in the population being studied.

Studying That Suits You

Use AI to generate personalized quizzes and flashcards to suit your learning preferences.

Description

This quiz explores important concepts related to research methods, focusing on online surveys and personal interviews. Test your understanding of various techniques, advantages, limitations, and key considerations in designing effective surveys. It's essential for students of research methods and statistics.