Planning a Study What question are you trying to answer? How are you going to collect your information? Deciding on what and how to measure?
2 Types of Studies Observational Study Experimental Study
Vocabulary Measuring What? Units Experimental Units Subjects Participants Various Variables Explanatory (independent, x) variable Response (dependent, y) variable Confounding variable
USA Today Headline “Prayer can lower blood pressure” “Attending religious services lowers blood pressure more than tuning into religious TV or radio, a new study says.” Confounding occurs if effects from participation of religious activities AND social support cannot be separated. Amount of social support (confounding variable) is RELATED to attending religious activities regularly (explanatory variable). Amount of social support (confounding variable) AFFECTS blood pressure (response variable). How you participate in religious activities (explanatory variable) influences your blood pressure (response variable) ?
Variables AFFECTS RELATED TO
Observational Study Observing the behaviors of a sample from a population. The observer does not impose active treatments on units/subjects. Or using previously collected data to do statistical analysis.
Census--Observational Study The systematical collection of data on the entire population. When the population is large, it will be time consuming and expensive.
Sample Survey-- Observational Study A portion of the population is asked a question and the study is done based on their voluntary answers.
Real Surveys, Real People, Real Lies In Feb. 1995, the Washington Post added this fictitious question to its weekly poll of 1000 randomly selected respondents: “Some people say the 1975 Public Affairs Act should be repealed?” Almost half (43%) of the sample expressed an opinion, with 24% agreeing and 19% disagreeing!
Newsweek announced “A Really Bad Hair Day: Researchers link baldness and heart attacks.” The article reported that “men with typical male pattern baldness…are anywhere from 30 to 300 percent more likely to suffer a heart attack than men with little or no hair loss at all.” The report was based on an observational study conducted by researchers at Boston Univ. School of Medicine. They compared 665 men who had been admitted to the hospital with their 1 st heart attack to 772 men in the same age group (21- to 54-years old) who had been admitted to the same hospital for other reasons.
Characteristics of a well- designed and well-conducted survey Trained interviewers must be consistent with asking neutral, non-leading questions. An unbiased sampling should represent the population of interest.
Misleading Survey Question Do you watch cartoons? Do you still watch cartoons?
Population Random Selection Sample
Sampling Vocabulary Population of Interest who you wish to represent in your sample Sample those selected randomly to represent your population of interest Random sampling reduces bias when selecting those from your population of interest Sampling frame a “list” of population subjects
Sampling Methods Simple Random Sample (SRS) Stratified Random Sampling Cluster Sampling Systematic Sampling Multi-Stage Sampling Random Digit Dialing Self-Selected Sample Convenience Sample “Quickie Polls”
Simple Random Sampling From the entire population every possible grouping of specified size has same chance of being selected.
SRS Suppose your population consisted of 4 shapes and you wanted to have a SRS of size 2.
Suppose your population consisted of 5 shapes and you wanted to have a SRS of size 2.
Stratified Sample vs. Cluster Sample 1st divide population into groups (strata), then take a Simple Random Sample from each strata 1st divide population into groups (cluster), then randomly select some clusters and sample everyone in that cluster
Systematic Sampling & Random Digit Dialing From a list, divide into consecutive segments (every 50 names), randomly choose starting point (21st entry), then sample at that same point in each segment (21, 71, 121, 171, …) Sample that approximates a SRS of all households in US that have telephones with a specific exchange ( )
Multi-Stage Sampling “survey designers might stratify population by region of country, then stratify by urban, suburban, or rural, then choose a random sample of communities within those strata. They would continue to divide communities into city blocks (fixed areas) as clusters, and sample from the selected clusters.”
Self-Selected Sample--radio station call-in Convenience Sample--surveying folks in a mall who appear willing to talk to you “Quickie Polls”--hastily designed, poorly pre-tested, one night survey sample for evening news show
Sources of bias in surveys If a selection process consistently obtains values too high or too low, then BIAS exists. Selection Bias Non-response Bias Response Bias
Survey Questions Unnecessary complexity to question Misleading question Ordering of questions Ensuring confidentiality Anonymous survey
Experiment Subjecting the sample to a controlled treatment where one variable is altered. The objects on which the treatment is imposed on is called experimental units (human subjects). Explanatory variables are called factors and specific values of the explanatory variable are levels.
Experiment Only long term well-designed experiments can be used to imply CAUSATION between the explanatory variable and the response variable. [Surveys can NOT!] Often the media misinterprets results from observational studies reporting “proven” links when in statistics we only have shown evidence of a relationship. Human subjects in an experiment are volunteers.
Is weight training good for children? If so, is it better for them to lift heavy weights for a few repetitions or moderate weights a larger number of times? 43 volunteers 14--Heavy load group 15--Moderate load group 14--Control group Randomly assign to a group
Designing a Good Experiment Randomization--randomly assign subjects to treatment and control groups Control—groups or blocks Replication--consistency R-C-R “Differences in the response variable between groups, if enough to rule out natural chance variability, can then be attributed to the manipulation of the explanatory variable.” This will allow determination of cause and effect.
Randomization--Crucial “Researchers randomize to reduce the likelihood that the results will be affected by confounding variables and other sources of bias.” Randomize Type of Treatment Randomize Order of Treatment
Control Groups are used to control for UNKNOWN variability! Control group--receives standard/traditional treatment OR Placebo (sham) group-- receives no treatment but subjects believe they are receiving treatment Single-Blind Double-Blind
Designing an Experiment with Control in mind Block Design--”divide units into homogeneous (similar) groups (called blocks) and each treatment is randomly assigned to one or more units in each block.” Matched-Pair Design-- ”assigned either two matched individuals (identical twins) OR the same individual (repeated measure) to receive the different treatments” This controls for KNOWN variability.
Replication When administering the treatments, be consistent with each procedure for each subject/unit.
Quitting Smoking w/Nicotine Patches Recruited 240 smokers (volunteers) at Mayo Clinic from 3 large cities Randomly assigned 22-mg nicotine patch or placebo patch for 8 weeks. All attended counseling before, during, and after. Double-blind (neither volunteers nor nurses taking measurements knew type of patch) After 8-wk (1 yr), 46% (27.5%) of nicotine patch group quit smoking and 20% (14.2%) of placebo group quit.
Gathering Data Experimental Design Observational Study