Research Exam 1 Study Guide
Research Exam 1 Study Guide 3163
Popular in Industry Research
Popular in Communication Studies
This 16 page Study Guide was uploaded by an elite notetaker on Sunday October 11, 2015. The Study Guide belongs to 3163 at Ohio State University taught by Zheng Wang in Fall 2015. Since its upload, it has received 69 views. For similar materials see Industry Research in Communication Studies at Ohio State University.
Reviews for Research Exam 1 Study Guide
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 10/11/15
Research Midterm Study Guide Uses of media research data Programming decisions cancellations or renewals of shows selection of syndicated shows formats for radio stations Factor s on programming decisions 0 Size and demographic 0 Audience loyalty and exclusivity Q Anticipated rating 0 Effects the show will have on other shows buying and selling advertising time 0 Baseline figures are the rating and share 0 The price of ads are related to audience size buying and selling stations 0 Factors on buying and selling media companies decisions Market revenue Ratings and shares of company Audience demographics Economics Amount of competition Loca on Network affiliation policy making 0 Focus on what media does to people 0 Government regulation help set up this policy Research applications in the print media industrv The brief history see textbook pp340342 for a summary 1 Readership studies who reads the publication what items are read 2 Circulation studies the penetration levels of a publication in various markets pricing and delivery 3 Management studies job satisfaction management structure and the impact of ownership on content 4 Typography and Makeup studies the impact of design elements page layout illustrations text fonts 5 Readability studies how textual elements affect comprehension ex Cloze Procedure Wavs of knowing and the language of science Different ways of knowing OOOOO Characteristics of the scientific method Addresses questions of fact not value Its objective and systematic Empirical evidence Open to the public Cumulative and selfcorrecting Types of research might all be in the book since not explained in the notes Applied vs theoretical 0 Applied research deals with solving practical problems and generally employs empirical methodologies Because applied research resides in the messy real world strict research protocols may need to be relaxed For example it may be impossible to use a random sample Thus transparency in the methodology is crucial Q It accesses and uses some part of the research communities39 the academia39s accumulated theories knowledge methods and techniques for a specific often state business or client driven purpose 0 Theoretical research Social and health sciences interact with non empirical fields and use both observation based and preexisting concepts such as intelligence race and gender 0 In psychology the term quotconceptual definitionquot is used for a concept variable 0 A theoretical model on the other hand gives an abstract description of a given system Depending on the requirement a model may highlight one single aspect of the system and hide the other details A model normally contains enough formalism such that it is not ambiguous Because of this property theoretical experiments can be designed to estimate the performance of the system in different environments Given a good theoretical model I believe the performance of a strategy can also be evaluated 0 Quantitative vs qualitative Qualitative focus groups field observations intense interviews case studies and ethnography Quantitative surveys o Syndicated vs custom Concept 0 Abstraction formed from generalizing name color height Variable 0 Definition concept that has more than one value 0 Conceptual use other abstract terms to describe 0 Operational describe the precise exact way in which the variable is measured 0 Types 0 Levels of measurement 1 Nominal have no intrinsic numerical ordering of the values Gender color 2 Qrdinal have intrinsic numerical ordering but distance between value is not equal a lot or a little 3 Interval have intrinsic numerical ordering with equal distance between values but has no absolute zero temperature 4 Ratio have intrinsic numerical ordering with equal distance between values and has an absolute zero age height Hypothesis 0 What they are a tentative statement describing the relationship between two or more variables 0 Why we need it o How to write one o lV s and DV s 0 DV Consequenceoutput What the researcher wants to explain What is affected in an experiment 0 IV Causeinput What the researcher uses to explain DV What is manipulated in an experiment Strategies of research process L Define the problem l Examine the literature 00 Select key variables 4s Develop the research questions or hypothesis 5 Select the unit of observation and analysis Qualitative research Basic qualitative research methods 0 field observations 0 continuum of participant to observer 0 role may move along continuum during course of study 0 types observer no interaction observer as participant minimal interaction participant as observer often are acting as member of group participant fully acting as member of the group Observation process Setting Participants Events Acts Gestures Dialogue Field Notes Must be descriptive analytic autobiographical 0 focus groups 0 popular with academics and marketers 0 an indepth discussion on a specific topic led by a moderator 0 usually 612 people 0 participants are paid for the 12 hours they are there 0 Advantages Great for testing new ideas Conducted quickly Inexpensive Flexible Less inhibited than oneone interviews 0 Disadvantages Cannot generalize Does not answer quantitative questions Views can be swayed by a dominant participant Moderators can influence 0 intensive interviews 0 Types Structured specific sets of questions Open develop new questions Depth probing tell me more 0 Developing Interview Questions Interview questions are more specific than research questions Develop questions based on lit review observations and document analysis Do not ask leading questions Often helpful to start broad Modify based on pilot testing 0 case studies 0 as many data sources as possible are used to investigate a reallife phenomenon 0 Characteristics Particularistic focus on a particular phenomenon Descriptive final product is detailed description Heuristic helps people to understand what s being studied Inductive attempt to discover new relationships rather than verify existing hypotheses Sampling amp Sampling Error SE Basic concepts population entire set of elements sample a subset of the population census study of an entire population sampling selecting a subset of a population statistic summary value for a sample sampling frame is the source material or device from which a sample is drawn Probability vs nonprobability sampling Probability Sampling relies on random selection elements have a nonzero chance of selection lnvolves systematic selection procedures Follows the guidelines of mathematical probability Avoids possible researcher bias Likely to be representative Appropriate when generalizing with precision NonProbability Sampling Does not follow guidelines of mathematical probability Simpler easier Appropriate when there is no need to generalize with precision such as research employing the five qualitative methods Types of Non probability sampling what they are and advantagesdisadvantages 1 A convenienceavailability sample a Use readily available participants for the study mall intercept b Helpful in collecting exploratory information and pretesting 2 Avolunteer sample a Who willingly participate differ greatly from nonvolunteers and will therefore produce results with error i Volunteers tend to be different in many ways than nonvolunteers b Often used for experiments 3 A purposivequota sample a Participants or elements selected due to characteristics they possess b Individuals or items that do not meet the necessary criteria are not used i Purposive chosen knowing it doesn t represent the population but make selection based on the purposes sampling 20 students 10 Asian 10 caucasian ii Quota chosen to fill a predetermined or known percentage sampling 20 students 2 Asian 14 caucasians 4 A snowball Sample a Participants are encouraged to tell people they know about the study b Word of mouth is then used to recruit further participants c Sometimes an incentive is offered to those whose friends do participate Types of Probability sampling what they are and advantagesdisadvantages 1 Simple Random Sample a Every element of a population has an equal chance of being selected i without replacement vs with replacement b Often use a table of random numbers to generate the sample c Random digit dialing i A kind of simple random sample often used in telephone surveys 1 Randomly selecting 4 digit numbers to be added to the prefix in the city 2 Randomly choose a telephone number and then add a predetermined number to get a new number 2 Systematic Random Sampling a Start the sample at a random point b Every nth element of the population is selected i A population of 100 randomly start at the 44th element ii Decide on a sampling interval of 7 iii Sample is the 44th 51st 58th 65th elements c One possible problem Periodicity i The order of items in the population may bias the selection process 1 Sampling broadcast TV shows from a list with a sampling frame of 10 2 What is NBC aired every 10th show on the list 3 Sample would not actually be representative of all broadcast TV but just of NBC 3 Stratified Sampling a Used to guarantee that a specific subset of the population is adequately represented i Not possible with simple or systematic random sampling b Population is divided into subsets by a parameter of the population i The more parameters looked for the less possible elements there are to sample from c The random sampling is done on each subset d Two types i Proportionate stratified sampling 1 Sample is based on their size in the population 2 If 30 of the population is aged 1824 then so is 30 of the sample ii Disproportionate 1 A subset of the population is oversampled or overrepresented 2 If a radio station wants to target 2534 year olds in a survey they may sample more from this group Cluster Sampling a What is you don t have a complete list of the population and can t obtain one b Population is divided into groups or categories typically geographically based i Want to sample readers of a magazine in OH sample from each zip code in the state error created if one area is more or less affluent c Multistage Sampling i A kind of cluster sampling used often in national studies ii Cluster of counties or other geographical area selected iii Clusters are then narrowed by choosing a district iv Individual blocks are then chosen in the area v A convention such as third household from the NE corner is used and a household is sampled vi Often an individual within the household is then randomly chosen Decision on sample size Research method used Number of groups being studied Cost and time Number of dependent variables What has been done before Understand sampling statistics and what is SE Size of SE and sample size size of SE and confidence levels 0 Size of SE amp sample size 0 Size of SE is negatively related to sample As size increases standard error decreases Need to quadruple sample size to half SE 0 Sample size can be dependent on O 1Research method used 0 2Number of groups being studied 0 3Cost and time 4Number of dependent variables 5What has been done before O O Size of SE amp confidence levels n Size of SE is positively related to confidence levels The higher the confidence level the larger SE Q Is a 68 confidence level really good enough That still leaves you possibly having an incorrect statistic about 13 of the time 0 Most social science research is performed at the 95 confidence level 0 The 95 confidence interval is 196 standard deviations from the midpoint rounded to 2 standard deviations in textbook figure Put SE CL amp 0 together At a certain confidence level the confidence interval is your sampling statistic and the SE calculated at that confidence level From the lecture slide In our last example The Confidence Interval is 5000 49 o t0 5049 o that iS oo 54900 So we say we are 95 confident confident level that the true population of how many OSU students think they are sports fans fall into and 49 SE of our finding 50 from our sample that is 451 549 confidence interval An example of SE CL amp Cl With a sample of 400 and a 50 sta9s9c 68 Confidence Interval 47505250 95 Confidence Interval 451 O5490 99 Confidence Interval 43575643 With a sample of 800 and a 50 statistic 68 Confidence Interval 48235177 95 Confidence Interval 46545346 99 Confidence Interval 45465454 Quantitative Research My What is it Identify different types of surveys selfadministered face to face surveys and telephone 0 Selfadministered group mail emailonline Q facetoface at home in central location mall intercept Q telephone Advantages and disadvantages of each type 1 Selfadministered 0 Advantages anonymity reflection possible can produce detailed responses no interviewer bias can cover wide geographic area relatively inexpensive group quick 0 Disadvantages required literacy and computersinternet low response rate must be selfexplanatory little detail often provided won t fill out lengthy questionnaire no control over who fills out questionnaire or how it s filled out order of answering questions 2 Facetoface 0 Advantages greatest detail is possible further probing possible can be visuals honest responses due to rapport developed Q Disadvantages costly recruiting may be difficult potentially dangerous less control over interviewers greatest potential for interviewer bias ie faceto face survey on feminism limited sample ie mall intercept 3 Telephone 0 Advantages broad geographic reach able to reach a large range of demographics ie random digit dial RDD sampling rapid data collection can ask a lot in a short time doesn t demand literacy Q Disadvantages short amount of time no time to think about answers generating the sample unlisted telephone numbers led to Random Digital Dialing RDD currently helps due to lack of cell phone directories high refusal rate ie cell phone minutes a telephone survey on face food Types of survey questions How to write good survey questions 1 make questions clear be precise avoid specialized words or acronyms avoid double or hidden meaning ie how many TV shows do you think are a little too violent most a few or none need to be fair to all types of respondents 2 Keep questions short short concise items that will be understood are the besD 3 Remember the purposes of the research only include questions that are directly related to your research questions 4 Avoid doublebarreled questions avoid questions that asks about two or more things ie The ABC network has programs that are full and sexually explicit Do you agree 5 Avoid biased words or terms how the question is phrased could influence responses ie ln you free time would you rather read a book or just watch TV Where did you hear the news about the president s new economic program 6 Avoid leading questions avoid questions that suggest a certain response ie like most Americans do you read the newspaper everyday avoid double bind questions when any response is a yes to an implied premise ie Do you still do marijuana 7 Don t ask for highly detailed information few can accurately answer such questions ie In the past 30 days how many hours of TV have you viewed with your family 8 Try to avoid potentially embarrassing questions unless absolutely necessary such questions could inhibit responses possible tactic to avoid this is through a multiple choice question Types of closedended question 1 Dichotomous Response Response is one of two contradictory choices 0 Agreedisagree yesno Example Ohio s state legislature should allocate more funding to public state universities agree disagree no opinion 2 MultipleChoice Respondents can choose an answer from several options Example In general politicians tell the truth All the 1me Most of the 1me Some of the 1me Rarely Never 2 MultipleChoice cont 2a Should be inclusive include all possible responses Example What is your favorite television network ABC CBS NBC What if their favorite is Fox ESPN or TV Land Two possible ways to correct this question First corrected example What is your favorite television network ABC CBS NBC other please specify Second Corrected Example What is your favorite television network Please list only one 2 MultipleChoice cont 2b Responses should be mutually exclusive 0 Should be only one possible response per question 3 Rating scales Rating an item or list of items on an ordinal interval or ratio type scale Likert Scales Respondents rate their agreement with an item Example Television dramas contain too much violence Strongly Agree 5 Agree 4 Neutral 3 Disagree 2 Strongly Disagree 1 Semantic Differential Scales Two bipolar adjectives separated by typically seven spaces Pictorial Rating Scales A scale consisting of pictures to rate responses Example How do you feel about the current state of public television Rank Ordering Ranking items based on personal perceptions Example Here is a list of sports Please rank them in terms of their appropriateness for male or female participation Put a 1 next to the one you believe most appropriate for males 2 next to the one second most and so on Checklist question Often used in pilot studies to refine questions Example What things do you look for in a new television set Check all that apply Fillintheblank Used infrequently often used to test recall of things like commercials or media content
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'