Presentation on theme: "CHAPTER 6.1 Part 1. REVIEW Observational Study Observe Record Measure What is the difference between an observational study and an experiment?"— Presentation transcript:
REVIEW Observational Study Observe Record Measure What is the difference between an observational study and an experiment?
EXPERIMENTS Involve active data production Experimenters actively intervene by imposing some treatment in order to see what happens What are some examples of experiments?
SUBJECTS & TREATMENTS The individuals studied in an experiment are often called subjects. A treatment is any specific experimental condition applied to the subjects.
VARIABLES Response Variable: a variable that measures an outcome or result of a study. (Also referred to as the dependent) Explanatory Variable: a variable that “explains” or causes changes in the response variable. (Also referred to as the independent)
EXAMPLE Just a pinch between your cheek and gum “As the evidence of the adverse effects of cigarette smoke grew, people tried many different ways to quit smoking. Some people tried chewing tobacco or as it was called, smokeless tobacco. A small amount of tobacco is placed between the cheek and your gums. Certain chemicals from the tobacco are absorbed into your bloodstream and give the sensations of smoking the cigarette. This promotes studies on the adverse effects of smokeless tobacco. One study in particular used 40 university students as subjects. Twenty were assigned to the smokeless tobacco, and twenty were given a substance that looked and tasted like smokeless tobacco but did not contain any harmful substances.
CONTINUED The students were randomly assigned to one of the groups. The students’ blood pressure and heart rate were measured before they started chewing and 20 minutes after they had been chewing. A significant increase in heart rate occurred in the group that chewed the tobacco. Answer the following: What type of study was this? What are the response and explanatory variables? Which was the treatment group? Could the students’ blood pressures be affected just knowing they were part of a study? Experiment Explanatory – Chewed or not Response – students’ blood pressure & heart rate Tobacco group If a students’ blood pressure was impacted by the fact that they were in a study it would have impacted al the students not just the students in the smokeless tobacco group
EXAMPLE 2 Do student’s who take a course Via the web learn as well as those who take the same course in a traditional classroom? Nova Southeaster University studied the success rates of their students that enrolled in these two different type courses. The author’s of the study claimed that students taking courses online were “equal in learning” to students taking the same course in the classroom. Replacing college classes with online course would save the University money, so we should move to online courses. However, when evaluated further, their average test scores on the course material given before the courses started was 40.7 for those enrolled in the online classes, and 27.64 for those enrolled in the classroom. It is very difficult to compare students that are starting the course at different levels.
VARIABLES Lurking variable: is a variable that has an important effect on the relationship among the variables in the study but is not on of the explanatory variables. Example: WW2 Bombing Analysis After the war they studied the accuracy of strategic bombing with regressions. Some things made sense (different types of bombers had different accuracy levels, higher altitude meant less accuracy). But one variable was whether enemy fighters opposed the bombers, and this had the *opposite* effect from what anyone would expect (fighter opposition meant more accuracy). So what variable is still lurking? Cloud cover. If the weather was cloudy the enemy wouldn’t bother to send up fighters, and accuracy was terrible because in that era bombing depended on sighting landmarks on the ground.
LURKING VARIABLES What does it mean…… Going back to the Nova Southeastern University study….the student’s preparation for the course was the lurking variable. So, how much of the results can we say is from the course, and how much is for the preparation for the course? Ultimately, the fact that the group that did better before the courses started scored the same at the end, really says that the students in the online course preformed poorly in the class.
CONFOUNDING Two variables are said to be confounding when their effects on a response variable cannot be distinguished from each other. They may either be an explanatory variable or a lurking variable.
EXAMPLE Experiments that study the effectiveness of medical treatments on actual patients are called clinical trials. The clinical trial that made gastric freezing a popular treatment for stomach ulcers had this “one- track” design: Impose Treatment –› Measure Response Gastric Freezing –› Reduced Pain? The patients did report reduced pain, but we can’t say that gastric freezing caused the reduced pain. It might just be the Placebo Effect. A Placebo is a dummy treatment with no active ingredients. Many patients respond favorably to any treatment whether it is helping them or not.
CONTINUED The one-track design of the experiment mean that the placebo effect was confounding with any effect gastric freezing might have. A second trial, done several years later, divided the ulcer patients into two groups. One was treated with gastric freezing as before, and another received a placebo treatment. The results: 34% of 82 patients in the treatment group improved, but so did 38% of 78 of the placebo group. This shows the gastric freezing was no better than the placebo.
IMPORTANT POINTS It is hard to remove confounding, when only an observation is present One-track experiments often yield useless data because of the confounding and lurking variables. Experiments often offer better possibilities and results.