BSB Issues 150, Week 1
BSB Issues 150, Week 1 01:119:150
Popular in Biology, Society, and Biomedical Issues
verified elite notetaker
Popular in Biological Sciences
This 3 page Class Notes was uploaded by Wendy Liu on Thursday September 15, 2016. The Class Notes belongs to 01:119:150 at Rutgers University taught by Anthony Uzwiak in Fall 2016. Since its upload, it has received 640 views. For similar materials see Biology, Society, and Biomedical Issues in Biological Sciences at Rutgers University.
Reviews for BSB Issues 150, Week 1
Report this Material
What is Karma?
Karma is the currency of StudySoup.
Date Created: 09/15/16
Week 1: Scientific Method 12 September 2016 Biology, Society, and Biomedical Issues Professor Uzwiak Wendy Liu General Process of the Scientific Method What inspires the question? o Experience in the natural world o Data and ideas shared btwn members of the scientific community o Published info Develop and test hypothesis o Hypothesis – tentative explanations for an observed natural phenomenon A single observation may give rise to multiple hypotheses Must be rational and testable w/ experiments Experiments never 100% prove anything as being completely true They only support that there is a low probability of the observations being due to chance o Data – info gathered from observations using senses or tools that extend the capacities of our senses Qualitative data – categorical (red, orange, yellow,…) Quantitative data – measured with numbers (1cm, 1.5cm, 3cm,…) Data generated in experiments may support or contradict hypothesis o Objectively analyze the results & systematically evaluate the proposed explanations Inductive reasoning – set of observations are used to reach a general conclusion Deductive reasoning – extrapolation from general premises to specific results o revise hypothesis, alter prior assumptions as necessary Findings generated by experimentation are shared and analyzed by the scientific community o Peer review – 4-5 individuals in similar area of expertise review and add/make suggestions to your paper o Publish data + interpretations: contribute to the broader understanding of a phenomenon o Scientific theory – highest level of confidence in the correctness of an explanation Finding of experimentation may contribute to the broader society o Technology, policy, authoritative knowledge Experimentation Controlled experiment – observe the effect of systematically manipulating one or more variable o Control group – not subject to the manipulation o Experimental group – subject to manipulation Types of variables o Independent variable – variable that is manipulated Treatment – something that is administered to the experimental group; the actual manipulation of the IV Factor – controlled IV; a treatment variable whose levels are set o Dependent variable – variable that is measured; changes as a result of the treatment Characteristics of variables o Discrete – limited number of possible values o Continuous – infinite possible values Types of research o Observational study – collecting and analyzing data w/o changing existing conditions o Correlational study – measure naturally occurring variables Assess whether the variables are related Correlation ≠ causation o Experimental research - systematically manipulate independent variable and measure its effect on the dependent variable Can show cause-effect relationships Objective experimental design Experimental bias – experiment designed to favor certain outcomes over others Placebo effect – esp. for human subjects with drug testing – patient’s belief in the drug leads to beneficial changes that are not due to the drug itself, but the patient’s own expectations o Placebo – the “control” for drug testing – a sugar pill, which has no physical affect Double-blind experiments – research procedure in which neither the experimenters nor the subjects know who is in what group (control/experimental) eliminates possibility of placebo effect in subjects eliminates possibility of experimenters treating subjects from diff. groups differently Randomization – subjects randomly assigned (by chance) to an experimental group; creates homogeneous treatment groups w/o bias o Completely randomized design – subjects assigned to groups completely at random o Randomized block design (Blocking) – first divide subjects into homogeneous blocks, then randomly assign subjects within a block to an experimental group within the block Ex: subjects divided into male and female blocks; randomly assign males to an experimental group, randomly assign females to an experimental group, then perform experiments within blocks Don’t compare data across blocks, only within a block Replication – repeat an experiment: reduces variability in experimental results, increases statistical significance and confidence level of conclusions drawn from data Sample – subset of the population to perform a study on o Simple random sampling (SRS) – each individual chosen entirely by chance; each member of the population has an equal chance of being included in the sample o Stratified sampling – taking samples from each stratum/sub-group of a heterogeneous population, with the proportion of each stratum in the sample equal to the proportion in the population Ex: randomly select 50 male subjects, randomly select 50 female subjects; total 100 subjects with 1:1 male-female ratio, which is reflective of the Rutgers student population’s male-female ratio SRS within stratums so that the final sample is representative of the population o Multistage random sampling – series of SRS in stages of large to small areas Ex: a country is divided into states, SRS the states; within each state chosen, SRS for counties; within each county chosen, SRS for neighborhoods; can go even smaller (or larger) as necessary More for convenience (like door-to-door surveys) Statistical Inference Parameter – number describing a population Statistic – number describing a sample, computed from the data observed in the sample o The statistic may also describe the population if the sample is truly representative of the population (see objective random sampling methods as above) Variability – spread of the sampling distribution; how close/far apart the data points are o Larger samples generally have smaller variability