Week 1 Lecture: Methods
Week 1 Lecture: Methods COMM 1100
Popular in Communication in Society
verified elite notetaker
verified elite notetaker
HIST 121 Owen Bradley-History of the U.S. Since Reconstruction
verified elite notetaker
Popular in Communication Studies
This 4 page Class Notes was uploaded by Julia Belopotosky on Friday May 15, 2015. The Class Notes belongs to COMM 1100 at Ohio State University taught by Brad Bushman in Spring 2015. Since its upload, it has received 86 views. For similar materials see Communication in Society in Communication Studies at Ohio State University.
Reviews for Week 1 Lecture: Methods
Report this Material
What is Karma?
Karma is the currency of StudySoup.
Date Created: 05/15/15
Methods Lecture What is a theory 0 In science a theory is not a guess It s a well substantiated wellsupported and welldocumented explanation for observations Theoretical Stimulus amp Response Theories are composed of constructs or concepts that are linked together in some logical way 0 Theoretical Stimulus the conceptual variable that stimulates or causes an effect Independent variable 0 Theoretical Response the conceptual variable that is affected by the theoretical stimulus Dependent variable 0 Operational De nition a de nition of a theoretical concept that is stated in terms of observable operations procedures and measurements A precise statement of how you de ned and measured a theoretical concept Links unobservable variables to observable variables Lends itself to quanti cation Sometimes requires creativity Selfcontrol study cookies vs radishesdon39t eat 0 Independent Variable the variable the researcher controls it is quotindependentquot of the participant39s control It is a quotvariablequot because it has at least two levels or groups 0 Dependent Variable the variable whose value is presumed to quotdependquot upon the independent variable What makes a theory good Falsi able it can be tested and rejected Good theories have survived many tests 0 Scientists can never prove that something is true they can only prove that something is false Predictive accuracy it can reliably predict behavior 0 Internal Consistency there are no logical inconsistencies among the constructs Economy it only contains needed constructs Fertility it spawns other research Theory and Hypothesis Theories are general whereas hypotheses are speci c Hypothesis are derived from theories 0 Multiple hypotheses can be derived from one hypothesis Hypothesis is either accept or rejected it is much harder to reject a theory Methodological Approaches Correlational study Experiment Longitudinal study Survey research Content Analysis Randomization Random Assignment there is a different group for each level of the independent variable Assignment to groups is random if each person has an equal chance of being in each oroub experiments Random Selection a sample is randomly selected from a population if person has an equal chance of being selected Not enough resources to select everyone so you choose a representative subset surveys Correlational Study Random assignment is not possible for ethical or practical reasons The researcher cannot manipulate or assign determine the independent variable The researcher simply measures the strength of relationship between the two variables of interest 0 Ex Correlation of violent TV shows and aggression of children Strengths and Weaknesses 0 Strength can be used when random assignment and control of the independent variable are not possible 0 Weakness cannot be used to infer causality Conditions need to infer cause 0 Presumed cause and effect are correlated o The presumed cause recedes the presumed effect in time 0 Alternative explanations are eliminated Experiments The researcher has control over the procedures manipulating the variable of interest independent variable and holding all other variables constant Participants are randomly assigned to the levels of the independent variable Researchers treat participants identically Experimental Studies rule out alternatives 0 Y didn39t cause X the researcher controlled it o The researcher holds Z s constant or spreads out their effects evenly across conditions via random assignment Strengths and Weaknesses o Strengths can be used to infer causality o Weaknesses quotArtificialquot settings and measures Can only test shortterm effects usually less than 1 hour Longitudinal Study 0 A correlational research study that involves repeated observations of the same individuals over long periods of time sometimes even decades 0 Must have two different time periods time shows which came rst Strengths and Weaknesses o Strengths the presumed cause precedes the presumed effect in time Can look at cumulative effects and longterms effects 0 Weaknesses Random assignment not possible Very expensive to conduct Survey Research 0 Types of surveys interviewing telephone interviewing questionnaires Never trust conclusions from a single statistic that has no comparison Margin of error random sample of 1200 has a 3 margin of error regardless of the size of the population from which the sample was drawn if it is 79 margin of error says true percentage is between 7682 Strengths and Weaknesses o Strengths good for assessing public opinion sometimes the only method available 0 Weaknesses requires a representative sample Based on selfreports could be lied to Cannot be used to infer causality Content Analysis 0 A quantitative research methodology for studying the content of the communication TV shows movies video games Strengths and Weaknesses o Strengths Unobtrusive no impact on what is being studied Can study processes over long periods of time o Weaknesses Cannot establish effects Correlation Coef cient Ranges from 10 perfect positive correlation to 10 perfect negative correlation Sign gives direction of the relationship o 0 says the two variables are not linearly related 0 The value of the correlation coef cient indicates the strength of the relation 0 Few correlations are perfect quotsmallquot is i 1 quotmediumquot is i 3 quotlargequot is i 5 Getting Con dent with Effects Replication the same result is observed over and over again by different researchers Triangulation Employing different methodologies in order to examine a given phenomenon MeatAnalysis a literature review that combines the statistical results ie correlations from different studies conducted on the same topic