Extra Credit Problems must be handed in by the following lecture. They must be your own individual work. Each is worth Lecture 1/Chapter 1 a maximum of 4 points, and total maximum is 50 points. For problems involving survey variables, Benefits & Risks of Statistics data is available on the course website www.pitt.edu/~nancyp/stat-0800/index.html � Organization of course, defining “Statistics” along with instructions for downloading and use of a � Various designs to answer research question statistical package called MINITAB. � Identifying study design and details For problems involving internet articles or news reports, you must hand in a copy of the article or report � Completion of anonymous surveys itself and analyze it as instructed. Course Description Book/Course Divided into Four Parts This course introduces statistical reasoning to a diverse Finding Data in Life 1. audience. The main goal is the understanding of (scrutinizing origin of data) basic statistical principles so that the student can Finding Life in Data (summarizing data 2. understand research reports involving statistics and yourself or assessing another’s summary) its applications. Key parts of the course are sampling and study design, graphical and summary Understanding Uncertainty in Life 3. tools, probability, and basic inference techniques. (probability theory) Examples from a wide variety of subjects will be Making Judgments from Surveys and 4. featured. The approach will be more conceptual Experiments (statistical inference) than computational.
Definitions Example: Research study design � Variable : a characteristic that varies from one � Background : Researchers want to determine individual to another if sugar can cause hyperactivity in children. � Statistics: the science of principles and � Question: Is there a single best way to gather procedures for gaining and processing data information? (info about variables’ values for a sample) and � Response: using the info to draw general conclusions � Statistics: summaries of data (such as a sample average or sample proportion) Example: Sugar � hyperactivity? Design #1 Example: Sugar � hyperactivity? Design #2 � Background : To determine if sugar can � Background : To determine if sugar can cause hyperactivity in children, a researcher cause hyperactivity in children, a researcher could ask 5 people about their child’s sugar could ask 50 acquaintances about their child’s intake and whether he/she has ADHD. sugar intake and whether he/she has ADHD. � Question: Is this a good design? � Question: Is this a good design? � Response: � Response:
Definitions (ways to gather data) Example: Sugar � hyperactivity? Design #3 � Background : To determine if sugar can cause � Anecdotal evidence : Personal accounts, hyperactivity in children, a researcher could conduct usually by a few individuals selected an observational study: obtain a random sample of haphazardly or by convenience. 50 families, record children’s sugar intake and whether they have ADHD. � Observational study: Researchers observe � Question: Suppose this study shows that ADHD is what happens naturally in terms of variables more likely to occur in children with high sugar of interest. intake. Can we conclude sugar causes � Experiment: Researchers take control of hyperactivity? values of one variable to see how it affects � Response: values of another variable Example: Sugar � hyperactivity? Design #3 Example: Sugar � hyperactivity? Design #4 � Background : To determine if sugar can cause � Background : To determine if sugar can hyperactivity in children, a researcher could conduct cause hyperactivity in children, a researcher an observational study. Suppose it shows that could select 50 families; assign a low-sugar ADHD is more likely to occur in children with high diet to some, high-sugar to others. Then sugar intake. evaluate children’s activity. � Question: What other explanations are possible, � Question: Why is this an experiment , not an besides sugar causing hyperactivity? observational study? � Response: Other factors that can play a role: � Response:
Example: Sugar � hyperactivity? Design #4 Example: Sugar � hyperactivity? Design #4 � Background : To determine if sugar can � Background : To determine if sugar can cause hyperactivity in children, a researcher cause hyperactivity in children, a researcher could select 50 families; assign a low-sugar could select 50 families; randomly assign a diet to some, high-sugar to others. Then low-sugar diet to some, high-sugar to others. evaluate children’s activity. Then evaluate children’s activity. � Question: Should assignment of diets be done � Question: Should subjects know if they’re by asking for volunteers? getting low or high sugar levels? � Response: � Response: Example: Sugar � hyperactivity? Design #4 Definitions � Background : To determine if sugar can � The placebo effect is when subjects respond to the cause hyperactivity in children, a researcher idea of treatment, not the treatment itself. could select 50 families; randomly assign a � A placebo is a “dummy” treatment. low-sugar diet to some, high-sugar to others. � A blind subject is unaware of which treatment Then evaluate children’s activity. he/she is receiving. � Question: What would be an appropriate � The experimenter effect is biased assessment of placebo treatment to compare to a diet with (or attempt to influence) response, due to knowledge high amounts of sugar? of treatment assignment. � Response: � A blind experimenter is unaware of which treatment a subject has received.
Example: Sugar � hyperactivity? Design #4 GREAT NEWS! Town Confirms Oatmeal Can Help Lower GREAT NEWS! � Background : Suppose an experiment finds a Cholesterol 100 people in Lafayette, Colorado volunteered to higher proportion hyperactive among children eat a good-sized bowl of oatmeal for 30 days to see if simple lifestyle changes---like eating oatmeal---could help reduce randomly assigned to high-sugar cholesterol. (vs. artificially sweetened) foods. •After 30 days, 98 lowered their cholesterol. � Question: Can we conclude that sugar causes •With these great results, the people in Lafayette proved to hyperactivity? themselves that simple changes can make a real difference. � Response: Articles that appear at the end of a lecture are to be Example: Reviewing Study Design Issues read for homework, before the next lecture, and brought to class as reference for discussions. These Background : Consider Quaker Oats study. � are especially important for your understanding of the Questions: � material in the first part of the course. Was this an observational study or experiment? 1. Was the sample size large enough? 2. Was there a random assignment of treatments? GREAT NEWS! Town Confirms Oatmeal Can Help Lower GREAT NEWS! 3. Cholesterol 100 people in Lafayette, Colorado volunteered to Are the study’s results convincing? 4. eat a good-sized bowl of oatmeal for 30 days to see if simple Responses: � lifestyle changes---like eating oatmeal---could help reduce 1. 2. cholesterol. 3. •After 30 days, 98 lowered their cholesterol. •With these great results, the people in Lafayette proved to themselves that simple changes can make a real difference. 4.
Recommend
More recommend
