COMM 250 Agenda - Week 4 Housekeeping Team Pictures Team Issues Team Rosters (?) TP1, TP2 Lecture RAT 1 The Craft of Research Variables Research Questions & Hypotheses Operationalizing Variables
In-Class Team Exercise # 1 - Part I Paradigms & Paradigm Shifts Deliverable: list 6 major paradigm shifts in society (List Both the “Old” and “New” Paradigms) Additional Team Work Choose a team name TP1: Meet as a Team TP2: Peer Evaluation P&C
The Research Process Conceptualization Start with / Develop a Theory and Develop Hypotheses Planning & Designing Research Operationalize all Variables i.e., How will you measure each variable? (must be precise!) Methods for Conducting Research Plan the Study and Collect the Data Analyzing & Interpreting Data Run Statistics and Interpret Results Re-Conceptualization Back to the Drawing Board
4 Types of Variables Independent – influences another variable IV = “Predictor” variable Dependent – variable influenced by another DV = “Outcome” variable Control – variable one tries to control for Either: “keep constant,” balance across groups, or extract in the statistical analysis (aka a “concomitant” variable) Extraneous – variable not studied/interested in But it has some impact on the IV–DV relationship
4 Levels of Variables Nominal – simply categories for classification; the “numbers” assigned are meaningless Political Party, Gender, County Ordinal – levels are in a rank order US News’ Top 100 Grad Schools; Letterman’s Top Ten List Interval – numbers are meaningful, no “true zero” a 7 Point Scale, Fahrenheit or Celsius Temperature Ratio – numbers are meaningful, a “true zero” Height, Weight, Kelvin Temperature
RQs and Hypotheses RQs Open-ended, general When researcher is unsure or new to the area E.g.: “How does education level affect income? Hypotheses Predict a relationship When researcher knows an area, or has a theory E.g.: “The more education a person has, the higher his/her annual income.”
RQs use Variables; Hs use IV, DV Independent – influences another variable IV = “Predictor” variable Dependent – influenced by another DV = “Outcome” variable Sample RQ: “What is the relationship between education level and income? Sample H1: “The more education a person has, the higher his/her annual income.
RQs and Hypotheses RQs Open-ended, general When researcher is unsure or new to the area “How does education level affect income? Hypotheses Predict a relationship When researcher knows an area of has a theory “The more education a person has, the higher their annual income.”
Hypotheses Two-Tailed Hypotheses Non-directional – researcher predicts a relationship, but does not specify the nature “Education level is related to income.”
. One-Tailed Hypotheses Directional –predicts a relationship AND the direction of that relationship “The more education a person has, the higher their annual income.”
Operationalization Operational Definition Defines a concept in observable / measurable terms A scientist can propose/claim/offer virtually ANY operational definition of a concept – all he/she has to do is be able to defend it So operational definitions must be: Plausible (must make sense to most in the field) Measurable (must be specified in detail) Replicable (must be complete - so others can repeat)
Examples of Operational Definitions Good (Defensible): IQ = “score achieved on the Wechsler Adult Intelligence Scale” Poor (Indefensible): IQ = “how smart someone is” Good (Defensible): Educ Level = “highest grade completed” Poor (Indefensible): Educ Level = “total years in school”
In-Class Team Exercise # 1 - Part II First Do as Individuals, then produce a Team Version: 1) Create 2 Hypotheses (One 1-Tailed, One 2-Tailed) Relate the concepts: “regular exercise” and “health” 2) Create a specific, measurable Operational Definition of each concept 3) Which is the IV, which the DV? 4) Propose 2 (likely/possible) “Intervening Variables” Deliverable : a written version of the above
Correlation & Causality Correlation Two variables are related (as one varies, the other varies predictably) Causation 3 “Necessary & Sufficient” Conditions: Two variables must be shown to be related The IV must precede the DV in Time The relationship cannot be due to another variable (an “Intervening” or “Confounding” variable)