9/8 biostatistics class notes- Evolution of physical activity guidelineso 1978 exercise training to improve performanceo present daily physical activity to improve health- physical activityo exercise subset of physical activity includes things throughout the day- transportation- occupation- leisure- household/daily livingo exercise –  physical activity that is planned, structured, and repetitive, with the purpose of improving or maintaining one or more components of physical fitnesso physical fitness – a set of attributes that people have or achieve that relates …- total physical activity college vs. high schoolo still highly active person, distributed differentlyo in college: more commute, less household- is total movement associated with health related factors regardless of how much moderate-vigorous intensity exercise you perform?- Research processo Research questiono Hypothesiso Identify research designo Data collectiono Presentation of datao Interpretation of data- Learning objectiveso Define biostats and role in research processo Understand how samples are used to draw conclusions on populationso …- what is biostats?o Application of statistical principles to medical, public health and biological applications Collecting, summarizing, interpreting, information and making inferences…- Two typeso Descriptive biostats Enumeration, organization, and graphical representation of data from a sample Sets up second parto Inferential stats Draw conclusions from incomplete information Generalizing from the specific sample Use available infor in a sample to draw inferences about the population from which the sample was selected- Where do we get data? Sampling- Useful terminologyo Population Class of individualso Parameters Are numerical facts about he populationo Sample A subset of the populationo Sampling frame: The list from which potential subjects are drawno Statistics: Are numerical factors about the sampleo Variable Is something that can varyo Data The values you obtain by measuring variables- Overview on samplingo Population Unknown parameter Census – info on entire population- Impractical- Not always feasible- Depends on parameters being measuredo Sample Information used to describe/make inferences about that population If this is unbiased and free of sampling error, should allow goodconclusions to be made about population- Advantages of samplingo 1. Sampling makes possible the study of a large, heterogenous (different characteristics) population With a large population it may be almost impossible to reach all of them. Sampling enables…- Advantages of samplingo 2. Relatively cheap reduces study size to a reasonable numbero 3. relatively fasto 4. Can provide accuracy need good sampling method- disadvantages of samplingo 1. Biased, not representative, too small – conclusion may not be valido 2. large sample group and multiple subgroups – procedure becomes complicatedo 3. Advanced technical knowledge, requires someone familiar with sampling and bias bias can affect study results, need to identify sources of bias and prevent them from affecting results- sampling biaso 1936 presidential electiono landon vs. FDRo literary digest poll of 10 mil Americans, 2.4 mil responses Landon would be overwhelming winner of election FDR won with 62% of the votes… Shock to makers of poll The poll was discredited and discontinuedo What went wrong? Magazine had surveyed its own readers Registered automobile oweners Registered telephone users Landon supporters Didn’t represent the entire population Conclusions drawn weren’t true Sample wasn’t randomo Bias occurred because: Over sampling of Landon supporters Nature of over-sampling was related to voter preference The sample population wasn’t the same as the population of interest, all American voters!- What is a good sample?o Sample must be valido Validity depends on 1. Accuracy- free of bias- internal validity- study does what it says it was going to do 2. Precision- sample represents the population (external validity)- sampling processo 1. defining the population of concern population of interest- whomever you want to generalize resultso womeno childreno doctorso athletes- population definitiono note: the population from which the sample is drawn may not be the population we wantto center on- sampling:o influence representativeness sampling procedure sample size participation (response)- when might you sample the entire population?o Population is very smallo When you have extensive resourceso When you don’t expect a very high response rateo 2. specifying a sampling frame, a set of items or events possible to measure since I’s difficult to identify all people in a population, a sampling frame, which is a list or some other source, that identified individuals… sampling frame- ability to identify people within that sample- need to get hands on list of sample- be able to identify each person in some kind of wayo student id numbero username for spireo contact specific people- ex. league of women’s voters registration list might be a sampling frame practice;- joggers aged 40-65 how might you define a sampling frame?- Find group of people that run on regular basiso Marathono Clubso Stores- Ex. registration for boston marathon, active.com races, runner’s world magazineo 3. Specifying a sampling method for selecting items or events from theframe probability sampling vs. non-probabilty sampling probability sampling- every individual has equal probability of being selected simple random sampling- randomly sample everyone in population- one example could be taking a table of random numbersor lottery system is used to determine which individualsare selected- produces unbiased samples- advantages;o estimates are easy to calculateo simple random sampling is always an EPS design,but not all EPS designs are simple random sampling- disadvatages;o minority subgroups of interest in population maynot be present in sample- STRATA MUST be mutually exclusive and exhaustive- To compute overall population estimates, weights are required Stratisfied sampling- Draw sample from each stratum- Men & women- Advantages:o Good when population has high variabilityo Mix of people that