Please enable JavaScript.
Coggle requires JavaScript to display documents.
PRIMARY DATA AND QUESTIONNARIES (Important variables (Independent variable…
PRIMARY DATA AND QUESTIONNARIES
Questionnaires: all methods and techniques of data collection in which each participant of the research is asked to respond to the same set of questions in a predetermined order, meaning it is a standardized interview. Alternatively one could also term it an instrument.
tailored design method: precisely specified questionnaires on construction and usage.
descriptive or explanatory research. It will enable the researcher to identify and describe the variety in different phenomena and explanatory research will enable the researcher to examine and explain relationships between variables.
Uniformed response: when the respondent has insufficient knowledge or experience and might guess the answers. They then choose answers that are more socially desirable.
Higher response rate does not prove research findings impartial and lower response rate does not mean responses are biased.
Types of questionnaires
Self-completed: completed by the respondents and it is also referred to a survey. (internet, web, mobile, postal, delivery&collection)
Interview-completed: occur when the interviewer read the same set of predetermined questions to the interviewee and records his/her responses (telephone, face-to-face), higher response rate and more suitable for longer questionnaires.
Computer Aided Personal Interviewing: when the interviewer reads the question from a computer screen and enters the responses of the respondent in the Computer. Computer-aided telephone interviewing: is the same but then via telephone
Important variables
Independent variable: leads to changes in a dependent variable
Dependent variable: may change in response to changes in other variables
Mediating: transmits the effect of an independent variable to the dependent
Moderator: will afect the nature of the relationship
Factual / demographic variable: characteristics (age, gender, maritial status).explore how attitudes etc differ and to check whether data is representative for the population.
Attitude and opinion variables: these contain data that may have been needed to be answered before confucting the questionnaire. They are influenced by the context in which they are asked
Behaviour and event variable: influenced by context and contain information about what people did (behaviour) or what.
Collecting essential data (data requirements table)
1) decide if the outcome is descriptive or explanatory
2) subdivide each research question
3) identify the variables
4) establish a level of detail that is required from the data from each variable
5) develop measurement questions to capture data at the level required
investigative question
a question needed to be answered by the researcher in order t address each research question and meet the objectives.
Design questionnaire
it is valid when it enables accurate data and a questionnaire is reliable when the data is collected consistently
Four stages: 1) the researcher needs to be clear about the required data and designs a question 2) the respondent needs to decode the question in the way the researcher meant it 3) the respondent answers the question 4) the researcher decodes the answer in the way that the respondent intended.
Validity
Internal (measurement): the questionnaire actually represents what the researcher tried to measure
Content: measures whether the questionnaire provides the right coverage for the investigative question.
Criterion-related (predictive): focuses on the extent in which the questionnaire are able to make accurate predictions.
Construct: the extent to which the questions actually measure the presence of the construct what they were intended to measure (attitude scales).
Convergent: the overlap /correlation between the mentioned scales when differeent scales are used to measure the same construct
Discriminant validity: is applied to where different scales are used to measure constructs that differ theoretical
Reliability: will it produce consistent findings at different conditions.
Test the re-test: where correlating data are collected wit hthe same questionnaire as much under the same condition as possile. However a long time interval can cause differences in responses
Internal consistency: connects the responses to each question in the questionnaire with the other questions in the questionnaires. One statistic to use to correlate questions is Conbach's alpha where the scale items are combined a a scale to measure a particular concept
Alternative form: compares responses to alternative forms of the same questions or group. (check questions)
Question types
Open: useful if the researcher is unsure of the response (Exploratory)
List questions: respondent can choose from (non response may indicate uncertainty or lack of appropriate choice
Category: useful when analysing behaviour or attributes. The answer only fits in one category
Ranking: discover the degree of importance to the respondent
Rating: Collection opinion data, likert style: measuring how strong the respondent agrees or disagrees with a statement (the scale score for each case would be calculated by adding together the scores) . or semantic differential rating scale: the respondent must rate a single object/ idea on a bipolar rating scale with opposite adjectives (friendly, unfriendly), scale item: one rating question
Quantity questions: year of birth example, answer in numbers, selfcoded questions as the response is coded by the respondent
Matrix: contan a grid of questions that enables the researcher to record the responses to two or more similar questions at the same time.
Translating
Lexical meaning: the precise meaning of individual words
Idiomatic meaning: a group of words have a significant meaning to natives
Experiential meaning: the meaning that is derived from an experience
Grammar and syntax: meaning the ordering of words and phrases to form a proper sentence.
Source questionnaire means the questionnaire that is going to be translated and the target questionnaire is the translated questionnaire, both should be included