21.3: Data Collection and Experimental Design Designing a Statistical StudyIdentify the variable(s) on interest and the population of the study.Develop a detailed plan for collecting data.If you use a sample, make sure the sample is representative of the population.Collect the data.Describe the data, using descriptive statistics techniques.Interpret the data and make decisions about the population using inferential statistics.Identify any possible errors.
3Section 1.3Observational Study - the researcher merely observes what is happening or what has happened in the past and tries to draw conclusions based on these observations.The researcher does not influence the responses.The researcher observes and measures characteristics of part of a population.For Example: Motorcycle owners are getting older and richer.Data was compared based on income of motorcycle owners over a period of time.
4Section 1.3Experimental Study (or experiment) – the researcher manipulates one of the variables and tries to determine how the manipulation influences other variables.The group being manipulated is called the treatment group.The group not being manipulated is called the control group. Control groups are usually equal in size to treatments groups.All subjects are known as experimental units.Often, placebos are given to members of the control group to make them think they are receiving the treatment. Placebos are harmless, fake treatments.For Example: Drug trials for new medications
5Section 1.3 Observational Study Advantages: Disadvantages: Usually occurs in a natural setting (not a lab).They can be used when it is unethical or dangerous to conduct an experiment.They also can be used when the researcher cannot manipulate the variables.Disadvantages:Definite cause and effect cannot always be determined because other factors may have led to the results.They can be expensive and time-consuming.They also can have unreliable data when relying on recordings from the past.
6Section 1.3 Experimental Study Advantages: Disadvantages: Researcher has much more control (over who is in groups and manipulating the independent variable).Disadvantages:They may occur in unnatural settings such as labs and special classrooms.This can lead to the Hawthorne effectThey can also have confounding variables.
7Section 1.3 Data Collection Simulation: Survey: Mathematical or physical model used to reproduce the conditions of a situation or process.Often uses computers to collect data.Used for situations that are impractical or dangerous to recreate.Often save time and money.Survey:Investigates one or more characteristics of a population.People mostly administer surveys (internet, phone, mail, etc.)Surveys must be designed to eliminate bias.
8Section 1.3 Experimental Design Goal: Produce meaningful and unbiased results.Three key elements:ControlRandomizationReplication.
9Section 1.3ControlConfounding variable – occurs when an experimenter cannot tell the difference between the effects of different factors on the variable.For Example: You own a bookstore. You choose to remodel your store at the same time as a new college opens up down the road. Business increases and you can not tell if this is due to the remodel or the new college.The Hawthorne EffectSubjects who knew they were participating in an experiment actually changed their behavior. Therefore, the results of the research was affected.
10Section 1.3 Placebo Effect Subjects act favorably to a placebo when in fact they have been given a fake treatment.Blinding is implemented in order to reduce the placebo effect. Blinding is when the subjects do not know whether they have been given the placebo or are receiving treatment. Double blinding is when the experimenter AND the subject do not know whether the subjects have been given the placebo or are receiving treatment.
11Section 1.3 Randomization Randomization – process of randomly assigning subjects to different groups.Completely Randomized Design: subjects are assigned to groups through random selection.Randomized Block Design: subjects are grouped by similar characteristics (called blocks) and then assigned within each block.Matched Pair Design: subjects are paired together then assigned into the treatment/control group.
12Section 1.3ReplicationReplication – repetition of the experiment under the same or similar conditions.