External Validity refers to the generalizability of the findings of a study to a broader population. Here are some questions you can ask to evaluate the external validity of a correlational study:

• Sample Representativeness: Does the sample in the study adequately represent the population to which the researchers want to generalize the findings? Consider factors like age, gender, ethnicity, and socioeconomic status.
• Setting Specificity: Was the study conducted in a setting that is similar to the real-world contexts where the results might be applied? For example, a study on student stress conducted solely at a highly competitive private school might not generalize well to the broader student population.
• Demand Characteristics: Did the study design inadvertently influence participants’ behavior? For example, if participants know the purpose of the study, they might unknowingly alter their responses to conform to expectations.
• Researcher Bias: Were the researchers’ perspectives or expectations introduced into the study design or interpretation of results?

By considering these questions, researchers can increase the confidence in generalizing their findings from a specific study to a broader population.

Cell Phones Only – Whom Should Polls Call?

Estimating Cell-Phone-Only Households:

Here’s how Pew Research might estimate the number of cell-phone-only households:

• Dual Methodology: Combine traditional landline surveys with random digit dialing (RDD) of cell phones. RDD involves generating random phone numbers to include cell phones that may not be listed in directories.
• Weighting: Weight the results from landline and cell phone samples to reflect the actual proportions of landline-only, cell-phone-only, and mixed households in the population. This can be achieved using census data or other reliable sources.

Sample for Estimating Cell-Phone-Only Households:

A stratified random sample would be ideal. This involves dividing the population into subgroups (strata) based on factors like age, income, and location, and then randomly selecting participants from each subgroup. This ensures the sample reflects the proportions of these subgroups in the population and provides a more accurate estimate of cell-phone-only households.

Contacting the Sample:

• Landlines: Use traditional phone calls.
• Cell Phones: Utilize phone surveys with automated dialing systems (Interactive Voice Response – IVR) or live interviewers trained in conducting phone surveys via cell phones. Text message surveys or online surveys with phone verification could also be considered as supplemental methods.

Challenges:

• Non-response bias: People who don’t respond to surveys might differ systematically from those who do, potentially affecting the accuracy of the estimates.
• Cost: Reaching cell phone users can be more expensive than landline surveys.

By employing a combination of these approaches, Pew Research can gather data from both landline and cell phone users, leading to a more accurate estimate of the number of cell-phone-only households.