CAPI
CAPI (Computer Assisted Personal Interviewing) is a technique of questioning when the interviewer asks respondent in person and notes acquired answers to the electronic questionnaire displayed on the portable multimedia device (tablet, laptop, smart phone).
CATI
CATI (Computer Assisted Telephone Interviewing) is a technique of questioning when interviewer asks respondent over the telephone and notes acquired answers to the electronic questionnaire.
CAWI
CAWI (Computer Assisted Web Interviewing) is a technique of questioning using internet access when respondent fills in electronic questionnaire. Respondent can receive the form by e-email or find it on the website. This technique of questioning is not so expensive and time-consuming as CAPI and CATI. But there are few points which have to be taken into account when this technique is used, such as:
- Not everyone has a computer or internet access.
- Some respondents may worry about the data’s misuse and therefore may fill some (or all) of the answers wrongly.
- Results of questioning vary according to method of the selection of the respondents (custom database, online panel, random selection).
Correlation ratio
The correlation ratio informs us about the strength of the non-linear relationship between numerical variables. It can take on values from 0 to 1: The value 0 indicates independence of numerical variables while value close to 1 indicates strong dependence.
Categorization
It is a process of defining the categories which individual variables can assume and with which will be worked further in the research. We can create a category in the form of intervals for the age of respondents or other quantitative parameters (height, weight, etc.) – group more variations of the variable into one category – or, contrary, create separate category for each variation of the variable. Based on the detailed categorization, it is possible to categorize the first degree data, second degree data and data of higher degrees. Categories must be mutually exclusive, so that each respondent's answer can always possibly be included in only one category. Their number should be based on what is to be found and which additional analysis will be performed with acquired data.
Contingency coefficient
By the contingency coefficient we find out the strength of relationship between two qualitative variables. There are two contingency coefficients – Cramér’s and Pearson’s.
Correlation coefficient
The correlation coefficient is used to determine the strength of linear relationship between two numerical variables. It can take on values from -1 to 1:
- Negative value indicates a negative relationship when values of one variable increase and values of second variable decrease.
- Positive values inform about positive correlation when values of both variables increase.
- The more the value of the correlation coefficient is close to -1 or 1, the stronger the linear relationship between the variables is. The value 0 on the other hand indicates that the linear dependence between pair of variables was not detected.
Correlation
The correlation expresses dependence between numerical variables. According to the type of relationship between the variables we calculate the strength of correlation either by using the correlation coefficient (for linear dependence), or the correlation ratio (for nonlinear dependence).
Correlation Analysis
It is a process in which we find out the strength of the relationship between two numerical variables. By using the correlation coefficient we determine the strength of the linear relationship between the pair of variables, by using the correlation ratio we determine the strength of the nonlinear relationship between them.
Coding of responses
It is a process when a code (usually numerical) is assigned to the each question and its categories. It facilitates and speeds up data processing while using computer technology. Code can be assigned in advance to closed questions. In case of open questions it is necessary to study responses first, create mutually exclusive categories according to their content and then assign the codes to those categories.
Conversation
The most common data acquisition method of qualitative research, when experienced interviewer with psychological or sociological education speaks to either one (depth individual interview) or to a number of respondents. Conversation is also used in quantitative research, where larger number of respondents participate and the requirements for the interviewer are not that high.
Customer satisfaction
Information concerning customer satisfaction with product or service, which is important for the success of the company offering products or services on the market. Research can cover immediate satisfaction (immediately after the purchase, or utilization) or cumulated satisfaction (after a longer period or after a number of times the product is used).
Categorization of the second degree data
Comparison of combinations of two selected values and their variations and searching for similarities and differences between them (e.g. satisfaction of women with the product, dissatisfaction of women with the product, satisfaction of men with the product, dissatisfaction of men with the product). For more information on the process see categorization.
Categorization of the first degree data
Investigation of the frequency of individual values and their variables (e.g. number of women and men who filled in the questionnaire). For more information on the process see categorization.
Categorization of higher degree data
Comparison of combinations of higher number of values and their variables and searching for differences and relationships between them (difference with the satisfaction with the product of men with secondary education). For more information on the process see categorization.
Comprehensive investigation
Investigation to which all units of statistical population are subjected, which is time and financially consuming, often impossible.
Characteristics
Attributes of units in population, which can be judged based on:
- How many units in population have them:
- Common characteristics – common for all units, which should be added to the researched sample.
- Variable characteristics – vary for units in population, because they are the subject of research.
- The form in which they are expressed:
- Verbal characteristics – expressing the information with words and acquiring two (alternative verbal characteristics) or more alternatives (plural verbal characteristics).
- Quantitative characteristics – expressing acquired information in numerical form and expressing either the level of previously verbal characteristic (serial characteristic) or the value obtained through scale, measurement (measurable characteristic).