Week Four Notes
Week Four Notes PSY3213L
Popular in Laboratory Methods in Psychology
Popular in Psychlogy
This 7 page Class Notes was uploaded by Paige Lenczden on Sunday January 31, 2016. The Class Notes belongs to PSY3213L at University of Florida taught by Marina Klimenko in Winter 2016. Since its upload, it has received 29 views. For similar materials see Laboratory Methods in Psychology in Psychlogy at University of Florida.
Reviews for Week Four Notes
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: 01/31/16
Lecture 3-3: Content Analysis - Associated with archival data and analysis of its content - Can be applied to written or spoken data - Goal: find any systematic meaningful variations or patterns in the content which can possibly be used to predict future behavior - Allows researchers to transform qualitative data (words) into numbers and can be used to compute statistics - Ensor and Hughes 2008: o Analyze conversations between mothers and two year olds o Hypothesis: the connectedness of mother/child conversations will advance child’s social understanding Connectedness operation definition: speaker’s utterance is semantically related to other speaker’s previous turn A turn: utterances of one speaker bounded by another speaker’s utterances (Shatz and Gelman, 1973) or a significant silence (usually five seconds or more) o Conversational turns Unit of analysis is a turn Connected turn: speaker’s utterance is semantically related to the other’s interlocutor’s previous turn No. I bumped my nose. You did bump your nose didn’t you? But I think it’s all better now though, isn’t it? Initiation turn: speaker initiates a new topic and elicits a semantically response from the other interlocutor Do you want a slice of ham? Bumped my nose Failed turn: speaker’s turn is directed to the other interlocutor but fails to elicit a semantically related response Would you like a slice of ham now or will you wait? You’re so tired Piglet down Conflict turn: speaker’s utterance includes a prohibition, threat, or insult Unclear turn: speaker’s utterance was inaudible and, thus, is not transcribed sufficiently to code o Reliability of connectedness Interrater reliability (agreement among two independent coders on 20% of coded data Lecture 6-1: Coding Scheme - State your research question clearly - Unit of analysis – thing(s) or person(s) you are studying o i.e. gender stereotypes in commercials - Unit of Observation/Observational Period – level at which you collect your data (row in data table) o Will depend on your research question, and how you would like to measure it - Key Elements o Research hypothesis o Unit of analysis o Unit of observation – conceptual and operational definition Conceptual definition of gender stereotypes: as perceivers’ observations of what men and women do in daily life Operational definitions with measureable categories: Family roles: o Women are home-makes o Men are breadwinners - Methods to Quantify Observations and to Manage Observational Period o Does my observation occur multiple times? (frequency count) o Observational unit: one commercial o Is it either present or absent Interval method Nominal scale variable o Other methods to quantify: Frequency – for multiple occurrences of behavior in one fixed period (i.e. one commercial) Duration – to capture the length of behavior in one fixed period (i.e. one commercial) Relatively complicated; should only be used of duration plays a big part Latency – for repeated behaviors; captures time between the behaviors - Observational Units and Additional Codes o Good unit of observation is short, has a clear beginning and end, and fits the purpose of your study o Additional records: types of programs, central characters in TV commercials, age of characters, and time that commercials are aired Chapter Four Outline 4.1 Content Analysis in the Process: Simplified version of the process that goes into content analysis: 1. Formulating a research question and specifying a hypothesis 2. Providing theoretical, empirical, or logical rationale for selecting specific qualitative data and for conducting content analysis 3. Selecting a random sample (determine a unit of sample) 4. Transforming constructs into study variables a. Developing conceptual and operational definitions for study constructs b. Selecting a unit of analysis c. Selecting a unit of coding and a unit of context 5. Transferring coded data into a statistical program for statistical analysis Formulating Research Questions and Specifying Hypotheses - Sources of Qualitative Data o Qualitative data is non-numerical information that can come from a variety of sources Can originate from a carefully designed study that yields qualitative information, from a study designed by another researcher and for a different research purpose, or from public or private records of government or private institutions Example: 1997 study where researchers compared the life story narratives of 70 older adults to find out whether the stories of high generative adults differ from less generative adults, with generativity defined as having concern for the well-being of the next generation o Adult’s narratives are qualitative data that will have to be transformed into numbers to make it suitable for statistical analyses Example: observational study where 79 mothers and toddlers were observed interacting during a book reading session o Purpose: measure degree of intersubjectivity, or, the shared understanding between two or more people, to test the extent intersubjectivity has on predicting children’s emotion understanding two years later - Qualitative and quantitative data that was collected by someone else or has naturally accumulated is referred to as archival data o Example: U.S. Census Bureau is a government institution that collects quantitative data about the U.S. economy and its people This data is open to the public and can be used for academic purposes A researcher who chooses to use this type of data is said to be using archival, quantitative data collected by the government o TV ads can also be considered qualitative data because they carry meaningful messages These messages visually and verbally manifest our cultural and societal norms and beliefs, and, as such, can be systematically evaluated by a researcher to learn more about our society and its social issues Example: Mastro and Stern 2003 study – systematically evaluated portrayals of characters of different racial and ethnic background in TV ads to test the hypothesis that minority characters are portrayed in different contexts and with different frequencies than white characters o Authors argues that these differences are likely to have an effect on the viewer’s self-perception who are likely to identify more closely with the depicted characters of the same racial and ethnic background Theoretical, Empirical, or Logical Rationale for Content Analysis - In the Absence of Quantitative Data o Content analysis can be time-consuming and limited in the kinds of inferences it allows researchers to make The rationale to use qualitative data for content analysis could be as simple as the fact that qualitative data may be the only available data Example: historians use past documents to piece together historical events just because the nature of their inquiries limits them to using information left by the people of the past Example: program analysts, who are responsible for evaluating the merits of government or private programs, sometimes have to resort to the analysis of qualitative data (e.g. organizational documents, staff interviews) in order to be able to determine the quality of the program o Government Accountability Office: analyze the denial of health care coverage letters; an example of qualitative data - Theory and Empirical Evidence o Sometimes the choice to use content analysis is grounded purely in theory and/or empirical evidence Example: choosing to observe mother’s and children’s facial expressions for the study of intersubjectivity First reviewed all relevant theories and prior empirical research on the topic Learned that one way people establish shared understanding is through eye contact and by matching their partner’s expression - Converging Evidence o Content analysis is a good method to provide supporting evidence for findings that have been previously documented by need additional evidence o This design tips towards higher internal validity at the expense of lowering the external validity (its ability to generalize the findings to the general population) o In correlational research, the scale tips against internal validity and more towards the external validity o Since content analysis is especially useful for evaluating naturalistic and sometimes unstructured real life data, it can provide more evidence, and with a higher degree of external validity, for something that has been found in artificial settings or needs more real-life supporting evidence Example: 2010 Gerhards study of European citizens showed that the Turkish were least likely to accept “homosexuality as justified” in comparison to other Europeans This does suggest that the Turkish hold more negative attitudes towards gays and lesbians, but more evidence is still warranted o Finding more evidence can be done by expanding our understanding about Turkish culture and can possibly find roots of such negatives attitudes in Turkish TV programming and comparing their depiction of homosexuals in comparison to other countries o One can hypothesize that Turkish TV may completely lack gay or lesbian representation or may portray them in a negatively stereotypical fashion If this hypothesis is supported, it will provide more converging evidence for the conclusions of Gerhards In summary, what the purpose of content analysis may be, a researcher will have to conduct a thorough literature review on the topic and then demonstrate the suitability of content analysis to address the goals of the study Selecting a Random Sample and Transforming Constructs into Variables - Researchers must decide on who or what will be able to provide them with the information or the measure of the phenomenon - In a typical psych study, people are the population of interest and they are selected at random to provide information about the phenomenon - In content analysis of quantitative research, a sample is usually not your conventional sample of participants, and instead, this could be a collection of documents, interviews, TV commercials, etc. - In content analysis the phenomena of interest are typically concepts or abstractions that are non-tangible, have no physical properties, and cannot be directly observed. These are known as constructs - Coding Schemes: o Convert qualitative constructs into variables with scores o Process of developing a coding scheme: Provide a conceptual and operational definition of the construct(s) Conceptual definition – one that is expressed in theoretical/scientific terms Operational definition – expressed in concrete and measurable terms ( can be directly observed and measured) The two definitions should match as closely as possible Identify the unit of analysis What you want to analyze and from which to draw conclusions o Usually, but not always, the same as unit of the sample Example: each tested individual is the unit of analysis, since he/she has the score that is being identified in the analysis and is also the unit of sample Select a unit of coding and a unit of context These two terms set the boundaries of the portion of the unit of analysis that is to be coded to measure the construct of interest The boundaries can be less than or equal to the boundaries of the entire unit of analysis, but never can exceed them o Example: when examining the stereotypes in TV commercials, each commercial is a unit of analysis which will need to be broken down into smaller units of coding to be able to capture instances of stereotypes in every commercial The unit of coding could be the central character of a commercial that either behaves in a stereotypical or non- stereotypical way The unit of context will further specify how long a researcher has to observe that individual and/or in what context The information about the unit of analysis, the unit of coding, and the unit of context are all usually included in a coding scheme, along with the details/instructions about how the construct of interest is going to be operationalized and coded Once the construct is coded, these quantitative data can be transferred to an SPSS, a statistical software package, to be analyzed Typically, the first column is reserved for the unit of analysis with the scores (e.g. the people, commercials, or schools, or categories, if a study involves categorizing and counting the number of observations that fall within each category The subsequent columns represent the variables or measured constructs
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'