Question 1 used Chinese SPSS, question 2 used English version of SPSS. Noun explanations are interspersed in the operation steps.
1. Frequent network activities (multiple choice, 10 options)
1.1 Data introduction
Question 5 corresponds to the following fields in the data set:
1.2 Operation steps
(1) Define multiple response sets
The variable coding method here is the "dichotomy". The basic method of dichotomy is to design as many variables as there are options in the multiple-choice question. Each variable has two answers, "Yes" and "No". If this option is selected, the corresponding value is "Yes", otherwise it is "No", and the values are respectively assigned "1" and "0". If there are many options for multiple-choice questions, for example, there are more than 10 options, if you code according to the dichotomy, you should design more than 10 variables, which is too cumbersome at this time, so you should choose "taxonomy". The classification method also splits the multiple-choice questions into several variables, the number of variables is equal to the maximum number of options selected at the same time in Cases, and then the serial number corresponding to each case selected option is entered into the SPSS data file in turn.
After defining multiple response sets, frequency analysis and cross-tabulation analysis can be performed.
(2) Analysis task 1: Frequency analysis of types of network activities
Basic frequency analysis is used to grasp the data distribution of a single variable.
Percentage of cases: In information science, each data set is composed of a series of individual data; they are brought together to form a totality of evidence related to a specific problem. For ease of use, we organize them into cases, which are the sum of information related to a specific part of a project. The proportion of cases in the population is the percentage of cases. The percentage of chat cases here refers to how much this part of the chat accounts for the total individuals, excluding the individuals included in the check.
(3) Report analysis
N: number of responses 223
Percent: percentage
of responses Percent of Cases: percentage, the denominator is the sample size
7.6% = 17/223
31.5% = 17/54 (54 is the total
number of samples) from the response frequency distribution (online behavior ) And sample distribution. It can be seen that among college students’ online behaviors, emails are sent and received, documents are the most transmitted, the proportion of data search and chat is also relatively high, and the game reads relatively few newspapers.
Conclusion: Most college students go online mainly for study and work, and making friends is for entertainment.
(4) Analysis task 2: Analyze the distribution of multiple variables under different values (taking gender as an example) to
master the joint distribution characteristics of multiple variables, so as to further explore the mutual influence and relationship between variables. Contingency table Crosstabulation analysis is usually used, that is, frequency analysis under cross-grouping.
Pay attention to selecting the response here. The response percentage considers that there may be multiple choices for the case, and a single case under analysis may have multiple responses (multiple choices). If a case is selected, although the change trend is the same as the response percentage, The percentage sum is not 0, which is not intuitive to compare (if you don’t mind, you can choose both the case and the response).
(5) Report analysis
Group analysis, analysis of differences in the preferences of males and females.
Boys: send files (17.3%=19/110), network communication (15.5%), chat (14.5%), data search (12.7%), …., online shopping (6.4%), online games (5.5%),
girls ：Email (19.5%), data search (17.7%), sending documents, chatting (15.9%), …, online shopping (6.2%), online games (2.7%). It
can be seen that there are similarities in online behaviors between male and female students, and there are also comparative differences. Significant differences.
2. Analysis of college students' online hours
2.1 Data introduction
Question 4 corresponds to the following fields in the data set:
The sample sample is as follows:
The sample data here means "the sample is most likely to check mail at night>late night>afternoon>morning".
2.2 Operation steps
According to the requirements of the problem, use the classification method to set up dummy variables for analysis
Dummy Variable Lable Value
$Q4_1 1st period 1/2/3/4
$Q4_2 2nd period 1/2/3/4
$Q4_3 3rd period 1/2/3/4
$Q4_4 4th period 1/2/3 /4
(1) Define dummy variable group
(2) Frequency analysis
(3) Report analysis The
frequency is analyzed for the value of 4, that is, the time period in which emails are least likely to be sent. It can be found that the most unlikely time for college students to send emails is in the morning, followed by late at night.