Types of Data
Sampling Techniques
Bias
Obs. vs Exp.
Designing Exp.
100

Classify this data set: 

Dunks, Starbies, Aroma Joe's 

What is Qualitative Data? 

100

Explain the key difference between random stratified sampling and cluster sampling. 

  • Random stratified sampling divides the population into homogeneous subgroups (strata) before sampling, while cluster sampling selects entire groups (clusters) within the population.
  • In stratified sampling, random selection occurs within each stratum; in cluster sampling, entire clusters are chosen.
  • Stratified sampling ensures representation from all subgroups; cluster sampling is often used for geographically dispersed populations.
100

Define undercoverage bias and provide a real-world example of how it might occur in a survey or study. How could researchers potentially mitigate this type of bias?

  • Definition: Undercoverage bias occurs when certain groups in the target population are inadequately represented in the sample.
  • Example: A telephone survey that only uses landline numbers, potentially missing younger people who only use cell phones.
  • Mitigation strategies: Using multiple sampling methods, updating sampling frames, or employing statistical weighting techniques.
100

Define and differentiate between explanatory variables and response variables. Provide an example of each in the context of a specific research study.

  • Explanatory variable: The variable that is manipulated or observed to see its effect on the response variable.
  • Response variable: The outcome variable that is measured and expected to change based on the explanatory variable.
  • Example: In a study on the effect of exercise on weight loss, exercise duration could be the explanatory variable and weight loss the response variable.
100

Explain the concept of a block design in experimental research. How does this design help to control for extraneous variables? Provide an example of a study where a block design would be appropriate.

  • Definition: Block design groups experimental units into homogeneous blocks before applying treatments.
  • Purpose: To control for known sources of variability not related to the treatments.
  • Example: Testing crop yields with different fertilizers, using field sections as blocks to control for soil variation.
200

Classify: 

The number of coffee flavors at Dunks

What is Quantitative Discrete? 

200

Compare and contrast convenience sampling and voluntary sampling. Discuss potential biases associated with each method and how these biases might affect the validity of a study's results.

  • Convenience sampling: selecting easily accessible participants; potential for bias due to lack of randomness
  • Voluntary sampling: participants choose to take part; potential for self-selection bias
  • Both may lead to unrepresentative samples, affecting generalizability of results
200

Explain the concept of nonresponse bias. Describe a scenario where nonresponse bias might significantly affect the results of a study, and suggest two strategies researchers could employ to reduce its impact.

  • Definition: Nonresponse bias occurs when there are systematic differences between those who respond to a survey and those who do not.
  • Scenario: A health survey where individuals with poor health are less likely to respond, skewing results towards healthier respondents.
  • Strategies: Offering incentives for participation, following up with non-respondents, or using statistical methods to adjust for nonresponse.
200

What is a confounding variable? Explain how the presence of a confounding variable can affect the interpretation of research results. Describe a scenario where a confounding variable might be present and suggest how researchers could control for it.

  • Definition: A confounding variable is an extraneous variable that correlates with both the explanatory and response variables, potentially leading to spurious associations.
  • Effect on results: Can lead to misinterpretation of the relationship between variables of interest.
  • Scenario: In a study on coffee consumption and heart disease, age could be a confounding variable.
  • Control methods: Randomization, stratification, or statistical adjustment.
200

Compare and contrast randomized block design and completely randomized design. What are the advantages of using a randomized block design, and in what situations might it be preferable?

  • Randomized block design: Treatments randomly assigned within each block.
  • Completely randomized design: Treatments randomly assigned to all experimental units without blocking.
  • Advantages: Increased precision, control for known sources of variation.
  • Preferable situations: When there's a known source of variability that can be controlled through blocking.
300

Classify: 

The average distance Ms. Leighton walks with her puppy each day

What is Quantitative Continuous? 

300

Evaluate the strengths and weaknesses of stratified random sampling. Include how might this method be used to ensure representation of minority groups in a large-scale social study.

  • Strengths: ensures representation from all subgroups, can increase statistical precision
  • Weaknesses: requires knowledge of population characteristics, can be complex to implement
  • Application in social studies: stratifying based on demographic characteristics to ensure minority representation
300

What is response bias? Identify and briefly explain at least three different types of response bias that can occur in surveys or interviews.

  • Definition: Response bias refers to factors that influence how participants respond to questions, often leading to inaccurate or untruthful answers.
  • Types of response bias: Social desirability bias, acquiescence bias, extreme responding, central tendency bias, question order bias, etc.
300

Compare and contrast observational studies and experiments. Discuss the strengths and limitations of each approach, and provide an example of a research question that would be better suited to each method.

  • Observational studies: Researchers observe and measure variables without manipulation.
  • Experiments: Researchers manipulate variables and randomly assign treatments.
  • Strengths/limitations: Observational studies are more natural but can't establish causation; experiments can establish causation but may lack real-world applicability.
  • Examples: Observational for long-term effects of diet on health; experiment for testing a new medication's effectiveness.
300

Describe the matched pairs design and its application in research. How does this design differ from other types of block designs? Give an example of a research question that would be well-suited for a matched pairs design.

  • Matched pairs design: Each block contains two similar experimental units, one receiving each treatment.
  • Difference: More specific form of block design with only two units per block.
  • Example: Testing a new drug's effectiveness by comparing each patient's response to the drug vs. a placebo.
400

Classify: 

Your age

What is Quantitative Continuous? 

400

In what situations might random sampling be preferable to other sampling methods? Provide at least two examples and explain your reasoning.

  • Understanding that random sampling gives every member an equal chance of selection
  • Examples might include: large-scale national surveys, clinical trials requiring representative samples
  • Explanation of how random sampling reduces bias and increases generalizability
400

Compare and contrast undercoverage bias and nonresponse bias. How are they similar, and how do they differ in terms of their causes and potential impacts on research findings?

  • Similarities: Both can lead to unrepresentative samples and skewed results.
  • Differences: Undercoverage bias occurs at the sampling stage, while nonresponse bias occurs after sampling during data collection.
  • Undercoverage relates to inadequate representation in the sampling frame; nonresponse relates to failure to obtain data from selected participants.
400

In the context of an experiment, define 'experimental unit' and 'treatments'. How do these concepts relate to each other? Provide an example of an experiment, clearly identifying the experimental units and treatments.

  • Experimental unit: The object to which a treatment is applied in an experiment.
  • Treatments: The conditions applied to the experimental units.
  • Relationship: Treatments are applied to experimental units to observe their effects.
  • Example: In an experiment testing fertilizer effects on plant growth, plants could be the experimental units and different fertilizer types the treatments.
400

What does it mean for a result to be "statistically significant"?

  • Definition: A result is statistically significant when it's unlikely to have occurred by chance.
500

Explain the difference between Discrete and Continuous 

Discrete: count

Continuous: measure

500

Describe a real-world scenario where systematic random sampling would be particularly useful. Include in your answer how you would implement this sampling method in your chosen scenario.

  • A relevant scenario (e.g., selecting participants for a city-wide survey)
  • Clear explanation of implementation (e.g., using a city's voter registration list and selecting every 100th name)
  • Justification for using systematic random sampling in this context
500

A researcher is conducting a phone survey about political opinions. Discuss how undercoverage bias, nonresponse bias, and response bias might all potentially affect the results of this study. Provide specific examples for each type of bias in this context.

  • Undercoverage bias: Example - excluding those without phones or with unlisted numbers.
  • Nonresponse bias: Example - certain political groups being less likely to participate in the survey.
  • Response bias: Example - participants giving socially desirable answers about their political views.
  • Clear explanations of how each bias type specifically applies to the political opinion phone survey scenario.
500

Design a simple experiment to test the effect of a new study method on student test scores. In your response: a) Identify the explanatory and response variables b) Describe the treatments c) Explain how you would select and assign experimental units d) Discuss potential confounding variables and how you might control for them

  1. Key elements in the response:
    a) Explanatory variable: Study method; Response variable: Test scores
    b) Treatments: New study method vs. traditional method (control)
    c) Selection: Random selection from student population; Assignment: Random assignment to treatment groups
    d) Potential confounding variables: Prior academic performance, study time, teacher effects; Control methods: Stratified sampling, measuring and adjusting for study time, using the same teacher for all groups

500

Analyze some of the criticisms or limitations of the original study and subsequent replications.  

Mention of criticisms such as small sample size, lack of diversity, or alternative explanations for results.

M
e
n
u