Popular in Course
verified elite notetaker
Popular in Department
This 11 page Bundle was uploaded by Sami Mint on Sunday June 29, 2014. The Bundle belongs to a course at a university taught by a professor in Fall. Since its upload, it has received 89 views.
Reviews for Comm88finalstudyguidedolly.pdf
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: 06/29/14
Communication 88 Final Book Readings General Features of Survey Research 1 A large number of respondents are chosen through probability sampling procedures to represent the population of interest 2 Systematic questionnaire procedures are used to ask prescribed questions of respondents and record their answers 3 Answers are numerically coded and analyzed Large Scale Probability Sampling I Large samples are used to ensure precise estimates of the population I National studies require lots of time money and personnel therefore many surveys involve smaller samples drawn from state local populations I Conducted with nonprobability samples I Small scale surveys are used if you have a low budget or some specialized research purpose Systematic Procedures Interviews amp Questionnaires I Procedures are standardized in order to enhance the reliability of the data I Structured Interview objectives are specific questions are written beforehand I Unstructured Interview objectives are general discussion is wide ranging and spontaneous I Semistructured Interview specific objectives but interviewer has some freedom in meeting these objectives Quantitative Data Analysis Depend on whether a survey s purpose is descriptive seeks to describe based on certain characteristics attitudes experiences or explanatory investigate relationships between 2 or more variables Secondary Analysis the analysis of survey data by analysts other than the primary investigator who collected the data Uses amp Limitations to Surveys Surveys are used to describe and explain purposes o Offer the most effective means of social description Can address a much broader range of research topics than experiments can Disadvantages o Cannot causally explain variables direction of causality and third variables can be an issue o Less adaptable o Susceptible to reactivity Survey Research Designs Cross Sectional Designs Data on a sample cross section of respondents chosen to represent a target population are gathered at essentially one point in time Limited by the amt and accuracy of the information that the respondents can capably report Weakness don39t show direction of causality and not great for studying process of change 0 Contextual Designs I Sample enough cases within particular groups to accurately describe certain characteristics of those contexts 0 Social Network Designs I Focus on the relationships among social actors and the transaction flows occurring along the connecting links Longitudinal Designs Provides stronger inferences about causal direction and patterns of change The same questions are asked at two or more points in time 0 Trend Studies I Repeated cross sectional design I Cohort Studies Cohort people who experience the same significant life event within a specified period of time birth year Tracing changes across cohorts in repeated cross sectional studies 0 Panel Studies I Reveal which individuals are changing over time the same respondents are surveyed again amp again Drawbacks timely costly attrition rates and reactivity Key point panel surveys measure changes in individuals over time trend surveys track general social changes and cohort surveys gauge changes in age groups Key point survey researches seek to minimize error due to construction of the sampling frame coverage error sample selection sampling error securing sampled respondents nonresponse error amp data collection measurement error Face to Face and Telephone Interviewing Advantages over questionnaires clarification greater response rate greater response variability Face to Face Interviewing Response rate is high Allows for visual aids permits long and complex questionnaires Disadvantage cost amp difficulty of locating respondents Answering survey questions requires that respondents 1 Comprehend the question 2 Retrieve info requested from memory 3 Formulate a response in accord with the questions and the info retrieved 4 Communicate a response deemed appropriate Questions must be truthful relevant non redundant and clear Materials Available to the Survey Designer Open Ended and Closed Ended Questions Open Ended free response allows respondents to answer in their own words 0 Advantages may provide information that the researchers never thought about a wealth of information o Disadvantage coding this info is timely and costly amp may produce error CloseEnded fixed choice respondent chooses a response from those provided 0 Advantages require less effort and less coding 0 Disadvantages difficult to develop good questions Direct and Indirect Questions Direct Question there is a direct clear relationship between the question that is asked and what the researcher wants to know Indirect Question link between researchers objectives and question are less obvious ex Do you believe your co workers would mind having a woman as a supervisor Opposed to direct would you mind having a woman as a supervisor BECAUSE respondent might not be willing to answer about himselfwill most likely answer truthfully when talking about someone else Response Formats Ordinal Response Scales Likert Response 0 Strongly Agree Strongly Disagree 0 Excellent Very good Good Fair Poor 0 Very satisfied Moderately satisfied A little dissatisfied Very dissatis 0 Complete Confidence 1 2 3 4 5 6 7 No confidence at all 0 Fun x Boring Existing Questions If possible researchers should use existing questions developed by reputable sources Writing the Items Using Language Effectively Items must be unambiguous easily read sufficiently brief clear amp precise Items should be easy to read accurately if terms have multiple meanings the definition should be provided Use appropriate vocabulary for the target population Doublebarreled question two separate ideas are presented together as a unit o Ex What factors contributed to your decision to marry and have children what if you are married but don39t have kids Leading questions suggest a possible answer or make some responses seem more acceptable than others leading people to respond in a certain way o Ex How often do you smoke marijuana OR do you agree with many senators that taxes should be increased The Frame of Reference Problem A frame of reference is what the respondents are thinking about referencing when they answer a question o Ex how satisfied are you with UCSB people who check relatively satisfied can be choosing that one because of social reasons scholarships personal reasons etc o You can fix this by asking a follow up why you do feel this way Funnel Sequence moves from a general question to progressively more specific questions o Avoids the possibility that asking specific questions first will in uence responses to more general questions Inverted funnel sequence begins with most specific question and ends with most general o It can ensure that everyone is thinking about the same points or circumstances before expressing general opinions Response Bias Problems Social Desirability people may respond to questions based on how they should be responding rather than how they actually feel o Fix this indirect questions careful placement and wording of sensitive questions assurances of anonymity Acquiescence Response Set the tendency for respondents to be very agreeable Position Biases o Primacy effect most likely to pick the response options listed earlier o Recency effect when you hear the responses people are more likely to list the answer that was listed last Key features of experiments manipulation amp control The Logic of Experimentation True experiments research designs that contain the basic requirements to permit strong inferences about cause and effect Preexperimental and quasiexperimental do not have these requirements Basic Features Manipulated independent variable followed by a measured dependent var 2 groups one receives experimental treatment other doesn39t Except for manipulation both groups are treated EXACTLY the same Subjects are randomly assigned to a group Test of statistical significance to test the likelihood that the results could have occurred by chance Matching and Random Assignment To eliminate prior differences between groups researchers use matching 0 Match subjects on a particular trait IQ GPA and then randomly assign these people into groups Random assignment is the process of placing subjects in experimental conditions and random sampling is selecting research subjects from an eligible population Internal and External Validity Measurement validity do the operational definitions of a variable measure what they were intended to measure Internal validity a study is high on internal validity when it provides sound evidence of a causal relationship External validity question of generalizability can the results from this experiment be generalized to other populations from replication lab experiments are typically high on internal validity but low on external Sampling in Experiments Most experiments use readily available populations in their sample college students and use specific artificial lab settings9 low external validity Broadening the subject population increases external validity Staging Experiments Pretesting9 Subject Recruitment acquisition of informed consent 9 Introduction to experiment Random assignment 9 Manipulation of IV 9 Measurement of DV manipulation check9 Debriefing Experimental realism when an experiment is found to have impact on subjects and to seem real to them ludgment experiments subjects make judgments about stimulus materials and have little direct impact to subjects Mundane realism the similarity of experimental events to everyday practices The Experiment as a Social Occasion Reactive measurement effects responses due to subjects awareness of being studied Demand Characteristics When subjects enter the lab they have expectations about what will happen When they agree to participate in an experiment they often willingly do whatever is asked of them regardless of how dumb stupid and dangerous Demand Characteristics the particular cues in an experimental situation that communicate to subjects what is expected amp what experimenter hopes to find Evaluation Apprehension When subjects are apprehensive and nervous about being in an experiment They try to change their answers to satisfy the researcher Experimenter Effects Experimenter s expectations can skew the results of the study Minimizing Bias Due to the Social Nature of Experimentation Detecting Demand Characteristics o Simply ask subjects about their perception of the experimental situation Provide subjects with a false story about what the experiment is about Measure the DV in a different setting than the IV Keep subjects unaware that they are participating in an experiment o Ask subjects to adopt role of faithful subject Detecting experimenter biases o Doubleblind technique prevents the experimenter from knowing which condition a subject is in o Have a single experimental session that includes all subjects 0 Reduce amt of Contact between experimenter and subjects OOO Experimentation outside the lab Field experiments tend to be higher in external validity and mundane realism than lab experiments Threats to Internal Validity An experiment has internal validity when you can make strong inferences about cause and effect When we can39t separate the effects of the IV from possible 3rd variables we say effects are confounded 1 History events in the subject39s environment other than the IV that occur during the course of the experiment that may affect the outcome environmental like a recent political event a remark by a subject Maturation any psych or physical change that takes place within subjects that occur with the passing of time becoming hungrytired growing intellectually developing health problems Testing changed in what is being measured that are brought about by the reactions to the process of measurement tests become easier after practice subjects remember how to do the test etc Instrumentation unwanted changes in characteristics of the measuring instrument procedure Statistical Regression tendency for extreme scores on a test to move closer to the mean the second time a test is administered for example the assertive training program in extremely shy rated individuals Selection if there are differences in the composition of the control amp experimental groups ex AA program vs hospital treatment in alcoholics if economic well being was a factor the hospital treatment will do better because more people well off will choose that program Attrition the loss of subjects from the experimental groups differential attrition when the conditions of an experiment have different dropout rates Pre experimental Designs Design 1 The OneShot Case Study X 0 0 Some treatment is administered and then the group is observed tested to determine treatment effects 0 Provides no basis for comparing the findings with other observations Design 2 The OneGroup PretestPosttest Design O1X 02 o Observing a group introducing treatment observing group again 0 Provides a basis of comparison but still invalid Design 3 The Static Group Comparison Group 1 X 01 Group 2 02 0 Provides a set of data with which to compare the post treatment scores 0 Provides the scores of a control group which helps to control for history testing and statistical regression 0 Selection is a threat because there is no RA True Experimental Designs Design 4 The PretestPosttest Design RA 01 X 02 O3 04 Measures the experimental group before and after the treatment Suffers from the external validity threat of testing interacting with the IV called testing X interaction or testingtreatment interaction Design 5 The PosttestOnly Control Group Design RA X 01 02 Similar to the Static Group design but includes RA By eliminating the pretest the design becomes more economical and it eliminates the possibility of an interaction between pretest amp experimental Design 6 The Solomon FourGroup Design RA 01 X 02 O3 O4 X 05 06 More expensive but provides much more information Factorial Experimental Designs When 2 or more IV are studied in a single experiment they are referred to as factors Main effect the overall effect of the factor by itself bar graph Interaction effects the effect of one factor affects another factor line graph 0 When lines are parallel there is not interaction effect Quasi Experimental Designs When RA or control is not possible use quasi 1 Separatesample pretestposttest design RA 01 X X 02 2 Nonequivalent control group designs 01 X 02 O3 O4 3 Interrupted TimeSeries Design 01 O2 O3 O4 X 05 O6 O7 08 0 Example looking at how a gun control law affected homicides you could take data from the past 4 years amp later 4 years after the law was passed 4 Multiple timeseries Design 01 O O O X 0 O O 0 O2 0 O O O O O 0 Sources of Available Data 1 Public documents official records Vital statistics data on births deaths marriages divorces etc Death records Emile Durkhem s study Suicide US Bureau of the Census Census data is commonly available in 3 forms aggregate data on social units such as states towns and census tracts 1 percent and 5 percent samples of individual level data and individual records released 72 yrs after each census 2 Private documents Harder to obtain than public records but provides lots of insight Diaries letters notes 3 Mass media Newspapers magazines TV radio films Can analyze verbal and visual content 4 Physicalnonverbal materials Works of fine art clothing household items and various artifacts 5 Social science data archives Repositories of data collected by various agencies and researchers Advantages of Research Using Available Data Nonreactive Measurement 0 Reactive Measurement changes in behavior that occur because of subject39s awareness that they are being studied 0 Available data avoids this problem because the data is collected without the knowledge of those who produced it Analyzing Social Structure 0 Allows researchers to analyze larger social units Studying an Understanding the Past Understanding Social Change Studying Problems Cross Culturally Improving Knowledge through Replication and Increased Sample Size Savings on Research Costs Disadvantages of Research with Available Data Searching for amp Procuring Available Data Measurement of Key Concepts Data Evaluation and Adjustment Assessment of Data Completeness Content Analysis I A set of methods for analyzing the symbolic content of any communication 0 Selecting and Defining Content Categories 0 Defining the Unit of Analysis I Recording units units of analysis 0 Deciding on a System of Enumeration o Carrying out the Analysis Potentials amp Limitations of Field Research I Main purpose of conducting field research is to get an insider39s view of reality I Methodological empathy seeing things from the observed person39s point of View I It39s exible so it does will with rapidly changing situations I Recommended when it39s essential to preserve whole events when a situation is complex when the focus is on the relationship between the person amp the setting I Field research is less costly than experimental research in the lab I Ethical reasons like staging a riot are not appropriate to conduct in the lab it takes time dependent on observational amp interpretive skills of researcher difficult to replicate may lack generalizability Research Design amp Sampling I Begins with 1 broad substantive amp theoretical questions 2 a methodological approach based on observation in natural settings I Sampling revolves around convenience and accessibility I Theoretical sampling sampling broad analytical categories that will facilitate the development of theoretical insights Field Observation I Nonparticipant Observation 0 Someone who attempts to observe people without interacting o Structured when explicit and present plans for selectionrecording encoding is used can be used to measure the duration frequency of patterns 0 Unstructured the extent that these processes are implicit amp emergent allows for the discovery of behavioral patterns I Participant Observation 0 A researcher actively participates in the daily lives of people o To fully understand the shared meanings of a group Field Interviewing I Informants participants who are interviewed to gain more insight into a situation I Formal in depth interviews are used once the researcher has ben in contact with the informant for an extended period of time Stages of Field Research I Problems 1 selecting a research setting 2 gaining access to the setting 3 presenting oneself 4 gathering info in the field 5 analyzing the info and developing a theoretical scheme for interpretation I Select a setting9 gain access9 present oneself 9 take field notes 9 develop analysis 9 leave field Observer participant roles I Complete observer easier to play one role than to try to balance two I Complete participant access may be gained to info that would be withheld from an observer 0 Both avoids problem of reactivity I Covert research clarifying points by questioning those observed Membership roles I Peripheral members only marginally part of the settings they observe I Active membership assume functional role in the setting but retain sight of its temporary nature I Complete membership becoming fully immersed in the setting and attaining full member status Ethics standards of right amp wrong Research Ethics applying ethical principles to scientific research Data Collection and Analysis I Scientists have moral and ethical guidelines to provide honest and true results I Research misconduct fabrication falsification and plagiarism Treatment of Human Subjects I Harm o Ex Zimbardo s prison experiment 0 Costbenefit analysis assessing the full extent of costs and benefits o Researchers should inform subjects should screen out participants who may be harmed measures to assess harm afterward should be given I Informed Consent o Researches should give participants the opportunity to make voluntary amp informed decisions about whether or not to take part I Deception o Debriefing after deception is used researchers should tell participants what was really up with the survey 0 Use deception as a last resort considering alternative methods and sensitive ways to debrief I Privacy 0 Anonymity amp confidentiality
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'