Research Methods

I. Purpose/Problem       II. Conceptualization      III. Design/Data Gathering      IV. Results/Analysis      V. Implications/Conduction   I. Purpose/Problem What am I studying? II. Conceptualization different angles of my research idea? focus? interested in the difference maker, difference made or the process of making a difference most important concepts? distinctions/associations? III. Design/Data Gathering how will you answer the research questions? what methods will be used? IV. Results/Analysis what did you find? V. Implications/Conduction what do your results mean   Starting a Research V Explore a pool of experiences (your experiences/experiences of those helping us)   Questions to ask as you begin research: What is the condition?    X? What comes after it?     X—? What comes before it?    ?—X What comes inside of it?    X (over boxed ?) What comes inside of it?  (situationally)    ? (over boxed X) What is outside? (generally)     ? (over dashed boxed X)   1.   What is the condition?    X? focus of attention? (specific as possible) 2.   What comes after it?     X—? consequences of this condition? 3.   What comes before it?    ?—X a consequence of this condition? what produced this condition? 4.   What comes inside of it?     X (over boxed ?) what does this condition include/contain what is contained in this relationship 5.   What comes inside of it? (situationally)    ? (over boxed X) what is the immediate context for this condition? is there immediate context in which is occurs? 6.   What is outside? (generally)     ? (over dashed boxed X) what is the nature of the world in which this condition is found?   X?    ——— focus of attention X—?      ?—X   ——— association, cause/effect relationship X [?]   ——— distinctions ? [X]      ?[[X]]   ——— context  

3 parts of a concept label conceptual definition operational definition Label way to refer to something (public opinion, media use, people)   Conceptual Definition what is the verbal meaning attached to concept label? sometimes the label is enough, you don't need it. (intelligence: quantitative, analytical, verbal)   Operational Definition takes verbal meaning from conceptual definition and specifies the steps needed to measure/experience that concept. (develop steps for measuring conceptual definitions and devise a formula)   Operational definition should do a good job representing the conceptual definition, if it doesn't we aren't measuring what we thought we were   4 Levels of measurement Nominal Ordinal Interval Ratio   Nominal weakest form classifies people, things etc. label for a category refers to presence/absence of something equivalence: everything in that category has to be equal categories: exhausted and mutually exclusive  Ordinal rank on some dimension but not specific (horse race w/ no stop watch: 1st, 2nd, 3rd) non specific measurements, you know the order but you do not know the distance between has an order: any category can fall higher or lower to another categories: exhausted and mutually exclusive  Interval equal interval measurements  NO TRUE ZERO: when zero does not mean absence of the thing because you can go below it. (weather) Ratio DOES have a true zero, [[must have everything stated in interval]] ——— speed, distance traveled   Why should you start at the highest level of measurement?    you can start high and go low, but you cannot start low and go high   How will 4 levels of measurements be used?    chi squared data ——    pew data —— nominal/ordinal    analysis of variants (calculating mean) —— interval/ratio

Validity how well a measuring devise measures what is it supposed to. how well the operational definition matches the conceptual definition Reliability  a measuring is reliable if it consistently gives us the same answer. measured through stability, internal consistency and equivalency  is a necessary condition for validity = operational definition   Kinds of error in Operational Definitions: systematic random Systematic Error consistently not measuring what we think we are. operational definition doesn't represent conceptual definition well problem with validity Random Error  if random error in operational definitions is controlled, the measure (exam/survey) will be more reliable.     μ   = POPULATION MEAN x̄  or σ  = SAMPLE MEAN N   = POPULATION n   = SAMPLE   Population entire group or class group being studied Sample subjects of population that represents its population well Parameter value that describes a population Statistic value that describes a sample   Descriptive Statistics: [mean] statistical measure that takes one score that represents the information as a whole    Inferential Statistics: techniques that allow us to study a sample then generalize it to the population    —can help us generalize about the populations from which samples were selected    —can tell us how likely it is that the differences between samples reflect real differences between populations       Non Probability Sampling (cannot calculate sampling error, try to avoid)  chose who is available [doesn't necessarily represent population you are talking about] unqualified volunteer sample    purposive sample [survey those who relate to survey] quota sampling snowball sample [survey those who know people who have taken it]   Probability Sampling (can calculate sampling error)  random sampling: everyone has an equal chance of being selected to respond systematic random sampling stratified sampling multistage sampling   Central Limit Theorem the average/sum of a large bunch of measurements follows a normal bell curve even if the individual measurements do not          applies to distribution of sample means for ANY population          don't need large sample size before distribution is almost perfect (n=30)          can tell how much sampling error we have based on the relationship between samples and the normal curve

Central Limit Theorem applies to distribution of sample means for any population       Standard Deviation - average distance of scores from population mean Standard Error - average distance of sample means from population mean Z Score - # of SD's you are away from the mean Confidence Level - plus or minus 2 standard errors will give us a 95.44% confidence interval SS - sum of the squared deviations around the mean SD (population) = SS/N = (x-u)^2 SD (sample s) = SE (P) =

Holisti's  used with NOMINAL level data used to calculate percentage or agreement between two coders run on each item in a code book acceptable level of reliability: min 90% better 95% ^because formula doesn't consider agreement by chance^ M: # of coding decisions on which 2 coders agree N1: # of decisions by first coder N2: # of decisions by second coder                                                      RELIABILITY = 2M / (N1 + N2)   Scott's Pi used with NOMINAL level data % of observed agreement between coders % of expected agreement between coders (agreement expected by chance) the agreement you REALLY get divided by the max you COULD get if you take agreement by chance out.  acceptable level of reliability: min 85% better 90%                                        RELIABILITY = % observed agreement — % expected agreement / 1— % expected agreement             Recognizing Quantitative Data look for references to coders, a code book or interceder reliability look for a # (frequencies and percentages) in tables or text   Content Analysis + describes communication context examines changes in context over time answer research questions about message characteristics compare media content to "real world" establish starting point for media effects studies — content analysis alone cannot tell us effect content has on audience findings limited to categories, definitions used in that analysis can be time consuming/expensive    

Research is a never ending process    1 study will answer 1 set of questions.    How would you use the 6 types of box questions? choosing between two things understand all of the factors of an issue consume media responsibly  deepen your thinking   If we want to emphasize, we talk about # of people effected   If we want to deemphasize, we focus on the raw number of people effected   Use scale when talking about beliefs or values    lets audience express   Reliable ALWAYS = Valid Valid DOESN'T always = Reliable   There is ALWAYS chance for sampling error   Round to the hundredths place