EMPIRICAL METHODS

Pregunta	Respuesta
EVERY SCIENCE FIELD CONSISTS OF...	THEORY: hypothesis + source systems METHODS: explain and assess the theorical findings. Aim: to achieve valid and verifiable results.
EMPIRICAL is...	information collected in a laboratory or in the field, based on specific and sistematic analysis.
CHARACTERISTICS OF AN EMPIRICAL APPROACH	1. COLLECT INFORMATION 2. CLASSIFICATION IN ORDER TO COLLECT THE INFORMATION 3. INTERSUBJECTIVE TRACEABILITY
CHARACTERISTICS OF THE SCIENTIFIC APPROACH	1. Systematic data collection allows to formulate predictions and support findings with quantifiable values; 2. Proceed methodolically following a standardized set of rules. 3. distance to the object of investigation
INTERSUBJECTIVE TRACEABILITY	The empirical research is conducted INDEPENDENTLY OF THE PERSON and the personal preferences of the researcher. With the same systematic approach, two groups shouls come to the same results.
QUANTITATIVE VS. QUALITATIVE	QUANT: collect empirical observations on a numerical basis. - WIDTH QUALIT: give a complex and comprehensive description of a phenomenon. - DEPTH
OBJECTIVES OF EMPIRICAL RESEARCH	1. DESCRIPTIVE RESEARCH - systematic description of phenomena (percentages, average values, relative frequencies) 2. EXPLANATORY RESEARCH - if-then relations between two or more facts.
EMPIRICAL METHODS FOR COMMUNICATION SCIENCES	1. MEASURING TYPE AND ANALYSIS: quantitative, qualitative 2. DATA COLLECTION: surveys, content analysis, observation, physiological measurement 3. RESEARCH DESIGNS: experimental, non-experimental.
MEASURING IS...	the systematic assignment of numbers or symbols to the values of a variable. 1 (nemrical relative)- feminine (empirical relative) 2 - masculine
SCALES OF MEASUREMENT	1. NOMINAL: no logical sequence, values can only be distinguished by name. (gender) 2. ORDINAL: have two or more values and can be ordered (grades) 3.METRIC: values organized according to size and represent the multiple of a unit (height)
CLASSIFICATION OF VARIABLES	1. DISCRETE: nominal variable which has a finite number of possible values. Gender, school grades. 2. CONTINUOUS: there is an infinite number of values between any to values. (height, age)
AN INDEX IS...	a statistical device which summarizes a collection of indicators. In an index there is no description of individual indicators. It is itself a new variable and must represent the area of variables.
SCALE LEVEL VS. SCALE	Scale level: diversity of measurements Scale: sum of several individual measurements into one total value
TYPES OF SCALES	Thurstone: nominally scaled . Predefined statements. Likert: level of agreement or disagreement on a symmetric agree-disagree scale. Semantic differential: measures individual reactions in terms of opposing pairs.
3 QUALITY CRITERIA FOR MEASUREMENT	1. OBJECTIVITY: the results are independent from the researcher 2. RELIABILITY: achieving the same result after repeating an operation under the same procedures. 3. VALIDITY: measuring what it is supposed to be measured.
TYPES OF RELIABILITY (3)	1. TEST-RETEST: same people, 2 times 2. PARALLEL TEST: 2 subgroups, 1 time 3. SPLIT HALF: divide the test in 2 sets, compare results.
SAMPLING IS...	providiing information about certain characteristics of statistical populations
TWO WAYS OF SAMPLING	CENSUS: complete count PARTIAL SAMPLING: selection of a subset of units to represent a population.
ARBITRARY SAMPLING	the researchers select units of study just because they are available. Ex: street survey Representativity can be problematic
RANDOM SAMPLING VS PURPOSEFUL SAMPLING	RANDOM: any unit has an equal probability of being selected. PURPOSEFUL: units selected according to how useful they are for the research
RANDOM SAMPLING SIMPLE AND SYSTEMATIC	random extraction of elements from a limited population. Ex: lottery
RANDOM SAMPLING STRATIFIED	Division of a population into smaller groups which are formed based on shared attributes or characteristics. Ex: opinion of doctors about medical soap operas.
PURPOSEFUL SAMPLING CLUSTER	The total population is divided into groups (or clusters) and the researcher chooses by simple random a number of clusters. He analyzes all units within the cluster.
RANDOM SAMPLING MULTI-STAGE	the sampling is carried out in stages, in a row.
PURPOSEFUL SAMPLING QUOTA	quota sampling requires that representative individuals are chosen out of a specific subgroup. For example, a researcher might ask for a sample of 100 females, or 100 individuals between the ages of 20-30.
PURPOSEFUL SAMPLING TYPICAL	analyze only a few individual cases who are specially characteristic for all units of study in the population. Ex: autistic people
PURPOSEFUL SAMPLING EXTREME	elements that have characteristics with extreme values. Ex: heavy users of online products.
PURPOSEFUL SAMPLING CONCENTRATION PRINCIPLE	focus is on the part of the population where researcher suspects there are predominant elements. Ex.: Skiers - Bayern.
QUESTION TYPES	OPEN QUESTIONS: enable free-form answers and are mostly used for qualitative studies. CLOSED QUESTIONS: the respondents select from a predetermined set of answers.
CLASSIFICATION OF CLOSED-ENDED QUESTIONS	A) rating scale: more than 2 response options + defined number of levels. B) true-false: only two response options (yes/no or achievement/performance) C) multiple choice: more than 2 response options, useful to determine levels of knowledge.
RESPONSE BIAS HALO EFFECTS	biases that affect the sequence of questions in a questionnaire, influencing from one question to the next. EMOTIONAL/AFFECTIVE or CONTENT/COGNITIVE
RESPONSE BIAS TYPES (5)	cognitive and affective halo effects consistency and contrast effect social desirable non-opinions primacy-recency effect
RECOMMENDATIONS FOR CONSTRUCTING ITEMS AND AVOIDING BIAS	1. no generalizations 2. no double negations 3. no words with multiple meanings 4. no words which are familiar only to some people 5. only one topic per item 6. specify period clearly
TYPES OF SURVEY QUESTION	1. fact based questions 2. knowledge questions 3. attitude and opinion questions 4. behavioural questions
TYPES OF FUNCTION QUESTIONS	icebreakers funnel and filter transfer and resting verification credibility sociodemographic
STANDARDIZATION OF SURVEYS	STRUCTURED - fixed questions / same pattern UNSTRUCTURED - qualitative, depend on time and lenght (context) IN DEPTH - defined guidance and question sequence (semi-structured) GROUP INTERVIEW - encourages debate and discussion; mutual influence of the response behaviour.
SURVEY MODES	FACE-TO-FACE INTERVIEW TELEPHONE INTERVIEW WRITTEN INTERVIEW ONLINE INTERVIEW
PROCESS OF A SURVEY (7 STEPS)	1. Define the scientific problem 2. Divide the scientific problem into dimensions 3. Operationalize the research concept 4. Design questionnaire and conduct pretest 5. Conduct the survey 6. Analyze and illustrate results 7. Present results and answer the research question
CONTENT ANALYSIS IS...	a research method to retrieve information from documents or media products. AIM: a systematic, intersubjectively traceable description of content-related, form-related characteristics of messages.
QUANTITATIVE ANALYSIS IS...	a method to analyze large quantities of texts. Words are counted and the text is deconstructed into categories so everything that goes beyond the lines will not be taken into consideration.
CONTENT ANALYSIS AND SOCIAL REALITY (3 ASPECTS)	1. point out the CONTEXT of the reporting 2. analyze the motives and attitudes of the COMMUNICATORS 3. consider the possible effects on the RECIPIENTS
A CATEGORY SYSTEM...	defines the characteristics that will be measured by COUNTING. deductive or inductive
A CATEGORY IS...	the exact definition of what must be analyzed. They can be divided into SUBCATEGORIES. After constructing them, researchers assign VALUES to the characteristics.
REQUIREMENTS FOR CATEGORIES	1. COMPLETE - male vs female 2. PRECISELY SEPARATED - values mutually excluded
PROCESS FLOW MODEL OF CONTENT ANALYSIS	1. DISCOVERY CONTEXT: phenomena converted into scientific question 2. EXPLANATION CONTEXT: terms, theoretical construct, codebook, coding and analyzing 3. APPLICATION CONTEXT: content of the study is used and results are made public.
MAYRING CHARACTERISTICS OF THE METHOD	1. analysis deconstructed into single interpretation steps 2. traceable by other persons 3. intersubjectively verifiable 4. transferable on other objects 5. useable for others
MAYRING'S ANALYSIS STEPS (3)	1. summarize the content analysis: paraphrasing, generalization, reduction 2. explicative: use extra material to make unclear elements comprehensible 3. structuring: establish a category scheme, pretest, improvement and final evaluation.
CORRELATE TEXT MATERIAL WITH CATEGORIES 3 STEPS	1. DEFINE CATEGORIES: text elements explicitly defined in categories 2. STANDARD EXAMPLES: concrete text examples for each category 3. CODING RULES: rules are formulated to allow a clear correlation
PROBLEMS OF CONTENT ANALYSIS	1. polysemy of terms 2. draw conclusions on the recipient- or communicator-level 3. select a particular direction of evaluation - codebook
QUANTITATIVE VS QUALITATIVE CONTENT ANALYSIS	QUANT.: formal and content-related characteristics for large quantities of text QUAL.: a lot of details about a small amount of texts.

Siguiente

Descripción

Resumen del Recurso

Similar

	Creado por ESTEFANIA OLIVA hace alrededor de 8 años