PSYC-109_Notes_-_9.28.10.docx

PSYC-109 Notes ? 9/28/10 Partitioning Variance In order to understand research results, total variance must be partitioned into two parts: Total variance = systematic variance Systematic variance ? the part of total variability that relates to the variables we are investigating Systematic variance tells us whether the variation in one variable is related in a systematic was to variability in other variables under study But not all of total variability will show up as systematic variance There may be other factors affecting out outcome data ? factors that are NOT under investigation but nevertheless have an effect This is called error variance ? the portion of total variance that is unrelated to the variables we are investigating Error variance doesn?t necessarily mean we made a mistake in our research design and process although it could! Behavior is so complex that we can never investigate every factor related to it The next best thing is to have a way to partition the part of total variance that remains unaccounted for ? error variance Why is this so important? It?s like background noise and you need to control it as much as possible to ensure that our research is worthwhile Basic Probability We use descriptive statistics to describe and inferential statistics to understand variability Before we can begin examining the variety of research methods available to us as researchers, we need a basic understanding of inferential statistics Inferential statistics allows scientists to answer the question ?How likely is it that my finding would occur by chance alone?? In psychology, we want to know if our findings occurred because of our treatment or manipulation in experimental research (i.e. systematic variance) Or by chance! (i.e. randomness) We make probability statements frequently: I bet this will happen? It?s probably going to rain today Chances are? Because human variability is as wide as the weather, inferential statistics has the same, strong foundation in probability theory Learning an understanding the basic of probability theory will help you to better understand sampling distributions and normal distributions We will not go into ?strict? probability theory in this class, but you DO need to know the basics! What do we mean by probability? All attempts to define probability turn out to go around in circle: Probability can only be defined in terms of itself ?Probability is equated with likelihood, which is likened to certainty which depends on the chance of something happening which is basically its probability.? Psychological research is based in probability The primary question: Is the effect of X on Y so unlikely that it probably did not occur by chance? Psychologists assume that chance is responsible and seek to disprove that assumption Thus, science is disconfirmatory ? we ?know? by eliminating We are, in a sense, comparing what is probable and what is possible Probability (of an event or p) is The number of times a specific event (s) can occur out of the total possible number of events (N) The formula is easy and handy: p(e) = s/N This ratio can range from 0 (no chance) to 1.00 (absolute certainty) A ratio of .50 or ½ means that the event is just as likely to happen as not to happen