Empirical Methods

1. Scientific problem
2. Dividing into dimensions
3. Operationalization
4. Questionnaire design and pretest
5. Conducting the survey
6. Analysis and illustration of results
7. Presenting results and answering the research questions or problem

1. Fact based questions
2. Knowledge based questions
3. Attitude and opinion questions
4. Behavioral questions

1. Ice breaker
2. Transfer and resting
3. Filter and funnel
4. Verification

1. Face to face
2. Written
3. Telephone interview
4. Online survey

Systematic assignment of an amount of numbers or symbols to the values of a variable.the relations among the numerical have to correspond to the relations among the objects.

- Nominal: variables whose value cannot be organized in a logical sequence and which can only be distinguished by their name. Qualitative variable. ex.: gender.
They can be dichotomous: variables which have exactly two values
Polytomous: variables with more than two values.


- Ordinal: variable can be ordered or ranked in a natural sequence/ logical sequence ex.: school grades.

- Metric: organized according to size and can represent a multiple of a unit. Intervals have to be always the same ex.: hight.

Discrete: has specific values and cannot have values between these specific values.

Continuous: infinite number of values between any two values that a continuous variable can take.2

1. Objectivity: how independent the results are from the person who uses the instrument. Different researchers must achieve coinciding results.

2. Reliability: dependability of the measuring process. If we repeat the operation under same conditions we should have the same result.
A. Test-Restest: examines if the scores from one sample are stable over the time from testing with the same instrument. same test to the same people in different times. Problem: memory effects.
B. Parallel test: the group is divided in two comparable subgroups which will be administrated the survey that is applied at the same time. If we achieve a similar score in both then the measurement instrument is reliable.
C. Split half: divide items that measure the same construct into two sets. Divide the test in half.

3. Validity: wether an instrument actually measures what it is supposed to measure. Content related. Accuracy.

Reliable selection of individuals or units of study from the population.

Background: provide information about certain characteristics features of statistical populations.

1. Census: study of every unit, complete count.

2. Partial sampling: subset of units selected to represent all units to estimate characteristics of the entire population of interest.

1. Cost saving
2. Time saving
3. A census is practically impossible

1. Theory: contains the hypothesis and source systems of the respective subject area.

2. Methods: explain and asses the theoretical findings and considerations to achieve valid and verifiable results

Information collected in a laboratory or field based on specific and systematic analysis.

Descriptive: systematic description that permits prediction

Explanatory: methodologically, rules/standards, independence

1. Collecting experience

2. The system/ classification used in order to collect

3. Intersubjective traceability

1. Quantitative: empirical observation on numerical basis of a small number of selected characteristics. WITH

2. Qualitative: detailed statements about people. Comprehensive description DEPTH

1. Induction: from specific to a general theory. Method to develop theories. It begins without any theoretical concept as a background. REALITY-THEORY

2. Deduction: assumes the existence of a principle system, which then can be certified by observing the reality. THEORY-REALITY.

Selection of individuals or units of study because they are available, convenient and represent some characteristics. but the representativity is problematic.

Every unit of population has a chance of being selected

1. Simple and systematic RS: random extraction of elements from a limited population (ex: choose every nth element).

2. Stratified random sampling: division of population on some specific characteristics.

3. Cluster sampling: spatiotemporal defined conglomeration of elements of the population which form a structurally reduced representation of the respective population.ex: households, school classes etc.

Selection of units according how useful their analysis is for the research.

1. Typical sampling: units of study who are especially characteristic for all units of study.

2. Extreme case: elements that have characteristic with extreme values.

3. Concentration principle: part if the population where the researcher suspects the predominant part of there elements to be.

4. Quota: selection of units based on pre-specified characteristics so that the total sample has the same distribution of characteristics assumed to exist in the population.

5. Multistage: different sampling procedures are used.

Index: sum of several single indicators

Scale: measuring instruments that consist in several individual measurements which all cover similar variables and refers to the same theoretical construct.

Thurstone: several items whose values are summed up to get a total value.

Likert: at least 5 stages items which are combined to an index by summation.

Semantic differential: consists of mor or less 10 to 20 opposing pairs (used to determine meaning of a concept).

1. Open ended:
- free answers
- not provide response option
- for quality studies
- responses need category

PRO:
- higher complexity
- new aspects
- unexplored topics

CONS:
- time and effort
- fragmentation
- biased result

2. Closed - ended
- preset response options
- not overburned interviewee
- options must be real

- rating scale, more than 2 options

- true/false

- multiple choice

1. Cognitive/halo effect
2. Consistency - contrast effect
3. Social desirability
4. Non-opinions
5. Primacy recency effects

- avoid several meanings
- use words familiar to all people in the target group
- allocate only one topic
- avoid double negation
- avoid generalizations
- specify clearly a moment or period referred

Investigation method of communication science

Aim: to show influences in society by systematically analyzing media. What is the content of the messages and how are these understood and processed by the people.

- content of the constituent part of the text
- latent meaning structured
- distinctive single cases
- the elements that aren't displayed in the text

Adv.
1. Researcher is independent from the presence of the person
2. Paper does not react differently no matter when and how often it is analyzed

Dis.
1. Contents are predeterminated by documents
2. Interpretation is required
3. Further inquiries are not possible

- content analysis of big quantities of text
- deconstructing text into categories
- content expressed beyond text lines will not be taken into consideration
- numerical value need a valid scale of comparison
- results must be independent of the researcher

Objectives:
- context of the reporting
- motives and attitudes of the communicator
- possible effects in the recipients

Characteristics:

- systematic intersubjectively traceable analysis of large amount of text material
- detailed analysis and aims at an extensive category system

1. Summarizing: text is reduced to a comprehensible short version:
- Paraphrasing
- Generalization
- Reduction

2. Explorative: explains the text in a comprehensible way with additional material

3. Structuring: aim: filter out particular aspects to conduct a cross sectional study or to asses the material under specific criteria

Decides which text characteristics will be measured/analyzed by counting

1. Deductive: according to a theory
2. Inductive: looking the text the researcher decides what categories he is going to use
3. Mix forms

1. Coding
2. Controlling
3. Adapting code book

1. Discovery context
2. Explanatory context
3. Application context

Data gathered from existing information, documentation, literature. It is also scientific.

Systematic and intersubjectively traceable data collection is the basis for formulating predictions.

To proceed methodologically according to a process based on generally accepted rules.

Intersubjective traceability: empirical research is conducted independently of the person, personal preferences and researcher.

Experiment -> to identify cause-effect relations

Survey -> we collect attitudes and opinions among a population

Content analysis -> one can systematically describe media content

Observation -> determine people's actual behavior

Are measuring instruments that consist if several individual measurements which all cover identical or similar variables and refer to the same theoretical construct.

Is formed by summing up several single indicators.

Ex. Of categories:
Plurality of opinion
Independent media
Journalistic work
Legal framework
Institutional transparency
Production infrastructure

A reliable measure is measuring something consistently but you may not be measuring what you want to be measure. Reliability is related to the measuring procedure and validity to the degree of physical and content related accuracy.

Closed ended questions:
Pros:
- differentiated information about the value of a variable
- realization and evaluation are economical
Cons:
- respond tendencies can occur
- neutral response option is problematic

True-False questions
Pro:
- processing, analyzing and result gathering time is usually short

Cons:
- high percentage of random results
- few differentiated responses
- high tendency towards yes-answers

Multiple choice:
Pros:
- processing and analyzing data is economical
- giving random answers is less problematic because there are several response options

Cons:
- it could be difficult to find response options because all options need to have the same probability of being chosen

Pro
- cost effective
- fast delivery of responses is possible
- low personnel expense
- interview error are impossible
- multimedia presentations are possible
- large samples or census possible without additional expenses
- easy implementation of international projects
- automatic data entry
- ideal filtering possibilities
- the quality of response can be estimated based on the time spent to fill out the questionnaire

Cons
- only certain subgroups are appropriate
- return rate of responses is low
- self-selection of respondents, low motivation for participation
- displaying problems with different browsers
- questionnaire needs to be short
- no possibility to control the fill-out
- double participation is possible
- it is unclear who fills out the questionnaire
- condition: to be internet user
- in case of abandon by the respondent data gets lost