Exam 2 Remaining topics Flashcards
What happens when the sampling size increases?
Sampling error decreases
Therefore statistical power increases
True or False?
The best method to conduct a survey of elderly women is online.
False
Online surveys should be targeted at a young demographic.
Inferential Statistics
The body of statistical computations relevant to making inferences from findings based on sample observations to some large population
Univariate inferences
Using inferential statistics to make inferences and make a prediction or statement about our population
2 assumptions:
1. Sample must be drawn from population from which inferences are made 2. 100% response rate
T-Test
Most common in experiment; lets us compare variables and see how it differs between two groups; compare means of different groups
What is the average score of one group compared to the average score of another? Involves treatment groups and placebo groups--compare results of both groups over time
Evaluation research
Research undertaken for the purpose of determining the impact of some social intervention, such as a program aimed at solving a social problem.
Needs assessment studies
studies that aim to determine the existence and extent of problems, typically among a segment of the population.
Experimental designs
having complete control of the design, meaning the treatment group; cannot do this with surveys – there is potential for some contamination; not perfect
Quasi-experimental designs
non rigorous inquiries somewhat resembling controlled experiments but lacking key elements such as pre- and post-testing and/or control groups.
Qualitative evaluations
in terms of surveys, they are few and far between.
most experiments in social science research are quantitative. Interesting, but hard to work with.
What is the Nielsen survey?
A survey that uses a series of filter questions to gain information about your television viewing–if you pass a certain checkpoint, they put the device on your TV or ask you to record a diary
It gives you a couple of bucks to do it too!
True or False?
The marijuana case study was an online cross sectional study.
True
Why do we collapse variables? and what are the weaknesses of doing so?
We collapse variables when our hypothesis is too specific.
Collapsing variables is simply combining 2 different responses as one broader response
• Example:
o The question asked is “Are you Republican, Democratic, or Independent?”
o “Democratic” and “Independent” categories are collapsed
• Weaknesses
o Less specific, loses detail, loses variance, but may be necessary in testing hypothesis
HINTS (Health Information Trends Survey)
Four cross sectional surveys funded by the Cancer Research Institute that was publically collected and made available to researchers for secondary data analysis.
What was the Orphan Support Program, and who were its target baseline?
Evaluation study that advocated the education and public health with interventions in Malawi. The end goal of the program was to improve the health of HIV orphans.
Baseline comprised of two different sets of communities. During the first year, the first set of communities acted as a treatment group received the intervention while the second set served as the control and did not receive the intervention.
