Monday, March 9, 2020

Comm 307 Study Guide Essays

Comm 307 Study Guide Essays Comm 307 Study Guide Essay Comm 307 Study Guide Essay Comm 307 Midterm Study Guide Types of Hypotheses: ?tentative statement about the relationship between IV DV ?One-tailed ?predicts the specific nature of the relationship or difference ?EX: Females will talk more (higher word count) than males in mixed dyadic dinner conversations ? Two-tailed ?predicts significant relationship or difference, but does not indicate the specific nature of the relationship ?EX: There is a significant difference in the quantity of words used (talkatively) by males females in mixed dyadic dinner conversations ? Null (Ho)? predicts that groups will not vary on DV or that there is not a relationship between 2 variables ? Ho: r=0 H1: r=0 ?H0: male talkatively = female talkatively ?**you are testing the Null hypotheses Types of Research Questions: ?explicit question researchers ask about variables of interest ? Directional ?asks if there is a positive or negative relationship or a specific difference between two or more variables ?EX: Do females use significa ntly more words than males? ?Nondirectional ?when researcher asks if there is no relationship between two or more variables or a significant difference occurs between two or more variables EX: Is there a significant difference between the amount of words (talkatively) females and males? Variables: ?Any entity that can take on different values ?Concrete ?birth order (first born, middle child, baby) ?sex (male or female) ?Abstract ?age (a number that changes) ?level of public speaking anxiety (score) ?Relationships ?correspondence between two variables ?Correlation ?Positive, Negative, No relationship ?Strength and Direction Variables: ?Independent ?variable(s) that is (are) manipulated or changed ?we study the impact on the DV ?Dependent ?variables recorded or measured ?we study changes in DV **RESPONDER! ?Intervening Variables ?variable that intervenes between the independent variable the dependent variable ? Antecedent Variables ?must consider what happened previously ?a study exa mining the impact of conflict (IV) on marital satisfaction (DV) ?possible antecedent variables could potentially impact the results ? Variable Levels ?Nominal ?three rules: ?mutually exclusive ?equivalency ?exhaustive ?examples: ?biological sex (male vs female) ?heart attack (yes vs no) ?Ordinal ?three rules: ?mutually exclusive ?logical ordering of categories (more of something) ?categorical balance ?examples: socio economic status ?(lower, middle, upper) ?education level ?(high school, college, graduate) ? letter grades ?(A, B, C, D, or F) ?Interval ?variable where the values of the categories are classified in a logical order that represents equal distances between the levels within each category ?there is no absolute zero ?examples: ?likert scale ?strongly disagree, disagree, neutral, agree, strongly agree ? semantic differential/bipolar adjective scale ? good/bad, dirty/clean, strong/weak ? guttman or scalogram rarely used in comm research ?to ascertain belief ?Ratio ?variable where the values of the categories are classified in a logical order that represents equal distances between the levels within each category with the presence of an absolute zero ?examples: ?temperature, age, height, weight ?mass, blood pressure, speed, heart rate CALCULATING Mean Median Mode ?most frequently reported score ?Bimodal (2 different modes) ?No Mode Skewness ?positively skewed = tail runs to right of curve ?negatively skewed = tail runs to left of curve Kurtosis ?If Kurtosis is above 0, then distribution is peaked with short, thick tails If Kurtosis is below 0, then distribution is flat and has many cases in the tails Normal Distribution (Bell curve) ?mean, median, mode = same number = bell curve CALCULATING: ?Range ?distance between largest value (Xmax) smallest score (Xmin) ?range = Xmax Xmin ?Sum of Squares ?Variance ?the average distance of the scores for an internal or ration scale from the mean in squared units ?high variance = most of scores are away form the me an ?low variance = most scores are centered closely to the mean ?Standard Deviation ?summary statistic of how scores vary from the mean is expressed in the original units of measurement tells us on avg how far each score differs from the avg score ?why we care: ?for a study we might see a reported: (M=24. 5, SD = 2. 1) ? 68% between 22. 4-26. 6 ?95% between 20. 3-28. 7 Likert Scales ?one likert scale ordinal data ?multiple likert scales interval data ?strongly disagree-strongly agree Semantic Differential Scales ?determines differential/bipolar adjective ?one scale ordinal data ?multiple interval data Reliability ?accuracy that a measure has for producing stable, consistent measurements ?ex: does the watch work effectively? ?Tests for Reliability Test-retest ?same measure/different occasion ?Alternate ?different measure/same phenomenon ?Split-Half ?split to 2 groups/correlate scores ?Cronbach’s Alpha ?statistical test ?interpreting like grades ?a = excellent, b = good, c = r espectable, d =undesirable, f = unacceptable ? . 70 to get published ?Krippendorf’s Alpha ?Ways to improve reliability of an instrument ? item construction ?increase the length of the instrument ?improve the admin of the test Validity: ?degree to which the measuring instrument measures what it is intended to measure ? Types of Validity ?Face (content) Validity ?look at appearance of measure ?Criterion Validity ?look at how accurately new measure can predict well-accepted measures ? Construct Validity ?look at degree survey measures ?Threats to Validity ?Overlapping variables ?Measuring relationship satisfaction but actually measuring life satisfaction ? Interaction of Different Treatments ?intervening variable ?results form multiple treatments not from experiment ?ex: measuring improvement of public speaking in comm majors (its likely that other classes are contributing to success too) ? Interaction of Testing and Treatment ?when participants are sensitive or receptive to future measures of particular variable ? Hypothesis Guessing ?when participants guess what researcher is attempting to measure ? Evaluation Apprehension ?some individuals experience anxiety when they know they are being evaluated ? ex: white coat syndrome ?Experimenter Expectancies ?experimenter unknowingly influence a participants scores on a measure ? ex: accidentally telling the subject to quickly complete their measure (when they are measuring time as a ariable) ? Social Desirability Bias ?when participant changes a response to be seen in a better light ? ex: altering a survey response in case someone links the response to self Conducting Survey Research ?Prepare ?determine your question types (NOIR) ?use common sense putting survey together ?create clear instructions ?design your study ?complete pilot testing as needed ?Disseminate ?interview (face to face, telephone) ?self-administered ?mass a dministration ?mailed administration ?internet administration ?Improving Response Rates ?make survey easy to complete ?keep survey short use SASE (stamps envelopes to return) as needed ?include a good cover letter ?use multiple administration techniques Response Sets, Non-Response Bias Content Analysis ?a summarizing, quantitative analysis of messages ?conducting content analysis ?early stelps: ?theory rationale ?conceptualization ?operationalization ?detailed description ?unit of analysis: ?major phenomenon being analyzed within a study ? next steps: ?coding schemes (defined) ?sampling (determined) ?training pilot reliability Cohen’s kappa Coding Problems ?coding misinterpretations ?coder inattention ?coder fatigue

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.