Program design of field investigation;
(1) sampling design:
Sampling is to extract a part that meets the survey requirements from the social population and conduct a comprehensive study as a representative of the population.
random sampling
Simple random sampling method: refers to the direct extraction of n constituent samples from all units of a specific population.
Simple random sampling methods: direct sampling method, lottery method and random number table method.
Hierarchical random sampling: first divide all units of the population into several sub-populations according to one or several characteristics, and then extract a sub-sample from each category by simple random sampling or systematic random sampling.
Grouping random sampling: first, all the basic units in the group are divided into several mutually exclusive groups, and then randomly sampled according to each group.
Multi-stage sampling: a sampling method that divides the overall sampling process into two or more stages.
Firstly, the whole unit is divided into several groups according to certain signs, and according to the principle of randomness, several groups are taken as samples at 1 stage;
Then divide the 1 phase sample into several groups as the second phase unit of sampling, and select several groups from them as the second phase samples, and so on until the requirements are met.
Systematic sampling: a method of sampling at certain intervals.
Arrange and number all the units in the whole according to certain signs;
Then divide the number of units in the population by the number of units in the sample to get the sampling interval.
Then, the first sampling unit is randomly selected as the starting point of sampling in the first sampling interval;
Then, sample units are sequentially extracted according to the sampling interval until the number of sample units is sufficient.
Non-random sampling
Intentional sampling: the method by which investigators select and determine samples according to the research objectives and their own subjective analysis.
It is divided into impression judgment sampling and experience judgment sampling.
Impression judgment sampling is purely based on the subjective impression of the investigator; Empirical judgment sampling is to select samples according to the past experience of investigators and their understanding of the respondents.
Quota sampling: subdivide the population into several sub-populations according to some characteristics, and then distribute the total sample size according to the proportion of each sub-population.
Collective sampling: divide the population into several groups according to certain standards, each group is a sampling unit, and then select several groups from these groups by random method, and collect all the individuals in the selected groups into the sample of the population.
(2) Questionnaire design
Data collected by the questionnaire: personal background data, media usage indicators and communication effect indicators.
Answer forms: question type, choice question type and compound type.
Content: fact questionnaire, knowledge questionnaire, self-cognition questionnaire and attitude questionnaire.
Questionnaire acquisition methods: mail questionnaire and interview questionnaire.
(3) Statistical analysis
Descriptive statistical analysis: a statistical method to sort out and summarize the obtained data and show its distribution characteristics.
Inference statistical analysis: according to the information provided by the sample, the overall distribution characteristics are analyzed and demonstrated by probability theory.
Correlation analysis and causal analysis
Correlation refers to the non-causal relationship between variable changes.
Causality refers to the change of two related variables because of the change of one variable.