Example datasets

We will use several datasets to illustrate the use of our software.

Friendship Network Data

In this dataset, information on friendship network, alcohol use, smoking, the big five personality traits, and academic performance among college students is collected for three years in 2017, 2018, and 2019. The participants were undergraduate students and the sample size is $N = 165$. There were about an equal number of male and female students (45% vs. 55%) in the sample. The average age of the students was 21.64 ($SD$ = 0.85). The average GPA of the students was about 3.273 ($SD$ = 0.53) out of 5.

Information on two social networks was collected. First, each student was presented a list of all the students in the study and was asked to report his/her acquaintanceship with everyone else on the list, on a Likert scale of 0 to 4. Second, each student was asked to report whether the students on the list were their WeChat friends or not (WeChat is a popular social network platform in China). Therefore, there are two friendship networks: the first one is a real-life weighted acquaintanceship network (referred to as the acquaintance network) and the second one is a virtual unweighted social media network (referred to as the WeChat network). The two networks together can be viewed as a multiplex network. Data on personality, happiness, depression, and loneliness were also collected.

Attorney Network Data

The second dataset includes the cowork and advice network dataset from 71 attorneys from a law firm called SG&R in 1988. The dataset is available from the SIENA website. The first wave of network data will be used in the analysis in the current tutorial. The cowork information is collected by asking the company employees to select people who have worked on the same case with them. Additionally, information on an advice network is collected via asking respondents who they seek advice from at work. Several non-network attributes are collected alongside with the networks. From those, the office one works at (i.e., Boston, Hartford, and Providence) and years with the firm will be used for analysis.

Florentine Marriage Data

The dataset is from Breiger and Pattison (1986), where the social network indicates marriage alliances, and the non-network variables include (1) wealth, each family’s net wealth in 1427 (in thousands of lira); (2) priorates, the number of priorates (seats on the civic council) held between 1282- 1344; and (3) totalties, the total number of business or marriage ties in the total dataset of 116 families.