Do parasocial relationships with micro- and mainstream celebrities differ? An empirical study testing four attributes of the parasocial relationship

Figures & data

Table 1. Descriptive statistics of demographics.

	On following mainstream celebrities (N = 354)		On following Micro-celebrities (N = 208)
	frequency	percentage	frequency	percentage
Age	M = 19.14 (SD = 2.41)		M = 19.02 (SD = 2.16)
Gender
1 = Male	112	31.63%	89	42.79%
0 = Female	242	68.36%	119	57.21%
Time spent	M = 3.31 (SD = 1.67)		M = 3.14 (SD = 1.68)
1 = less than 5 mins	59	16.67%	37	17.79%
2 = 5 mins ~ 10 mins	74	20.9%	48	23.08%
3 = 10 mins ~ 20 mins	72	20.34%	33	15.87%
4 = 20 mins ~ 30 mins	48	13.56%	25	12.02%
5 = 30 mins ~ 1 hour	48	13.56%	24	11.54%
6 = more than 1 hour	51	14.41%	25	12.02%
Feedback received	M = 1.16 (SD = 0.53)		M = 1.36 (SD = 0.71)
1 = 0	307	86.72%	143	68.75%
2 = 1 ~ 5	40	11.3%	44	21.15%
3 = 5 ~ 10	1	0.28%	6	2.89%
4 = more than 20	4	1.13%	1	0.48%
Choice of platforms
QQ	113	31.92%	51	24.52%
Wechat	53	14.97%	46	22.12%
Tencent News	100	28.25%	27	12.98%
Sina microblog	269	75.99%	128	61.54%
Tik Tok	77	21.75%	67	32.21%
Youku	42	11.86%	10	4.8%
Zhihu	82	23.16%	40	19.23%
Kuaishou	6	1.69%	5	2.4%
Bilibili	137	38.7%	88	42.3%
Baidu Tieba	55	15.1%	22	10.58%
Douyu live stream	18	5.08%	24	11.54%
HUPU sports	11	3.1%	2	0.96%
Facebook	26	7.34%	7	3.37%
Twitter	35	8.89%	9	4.33%
Youtube	31	8.76%	9	4.33%
Others	51	14.4%	14		6.73%

Table 2. Zero-order correlations between the main variables and the control variables.

	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1. Age	1
2. Gender (girl = 0; boy = 1)	−.014	1
3. Time spent on MS	−.103	−.209**	1
4. Time spent on MC	−.077	−.021	.516**	1
5. Feedback from MS	−.023	−.018	.171*	.120	1
6. Feedback from MC	−.024	−.054	.190*	.205*	.474**	1
7. Reciprocity (MS)	−.069	.244**	−.009	−.044	.056	.052	1
8. Parasocial interaction (MS)	−.061	.161*	.050	−.092	.025	.101	.352**	1
9. Authenticity (MS)	−.004	−.048	.206*	.038	.120	.102	.068	.220**	1
10. Intimacy (MS)	−.065	.015	.242**	.091	.285**	.176*	.43**	.375**	.373**	1
11. Reciprocity (MC)	−.089	.178*	.018	.030	.248**	.183*	.49**	.392**	.210**	.509**	1
12. Parasocial interaction (MC)	.019	.198*	.118	.176*	.007	.250**	.278**	.461**	.018	.358**	.392**	1
13. Authenticity (MC)	−.052	.069	−.024	.067	.217**	.103	.002	.087	.321**	.145	.238**	.147	1
14. Intimacy (MC)	−.007	.197*	.069	.153	.277**	.245**	.307**	.535**	.207**	.610**	.643**	.447**	.265**	1

Note: N = 160; MS = Mainstream celebrities, MC = Micro celebrities. *p < .05, two-tailed; **p < .01, two-tailed.

Table 3. Comparing four attributes of parasocial relationships with mainstream and micro-celebrities.

	Mainstream celebrities		Micro- celebrities
Dependent samples (N = 160)	M	SD	M	SD	t
Reciprocity	2.07	0.76	2.35	0.82	−4.13***
Parasocial interaction	2.23	1.17	2.26	1.1	−0.257
Authenticity	3.54	0.78	3.32	0.86	2.817**
Intimacy	2.67	0.77	2.43	0.99	3.6***

Note: Results are based on 1000 bootstrap samples. *p < .05; **p < .01; ***p < .001

Table 4. Hierarchical regression analysis predicting the parasocial interaction.

	Test model Mainstream celebrity followers (N = 354)				Test model Micro-celebrity followers (N = 208)
	B	SE	β	t	B	SE	β	t
Constant	0.371	0.575	-	0.654	0.306	0.706	-	0.433
Step 1
Age	−0.007	0.024	−0.015	−0.299	0.019	0.032	0.038	0.578
Gender	0.116	0.118	0.050	0.981	0.165	0.143	0.078	1.156
Time spent	−0.017	0.035	−0.027	−0.497	0.070	0.043	0.113	1.647
Feedback received	−0.081	0.099	−0.040	−0.815	0.182	0.106	0.118	1.710
Step 2
Reciprocity	0.316***	0.078	0.217	4.057	0.296*	0.116	0.215	2.556
Authenticity	0.020	0.075	0.014	0.269	−0.142	0.080	−0.120	−1.785
Intimacy	0.481***	0.079	0.350	6.094	0.317**	0.094	0.283	3.378
R^2	0.229				0.260
∆ R^2	0.214				0.180
Adj. R^2	0.212				0.412
F(df1, df2)	13.836 (7, 326)***				8.776 (7, 175)***
∆F (df1, df2)	30.160 (3, 326)***				14.223 (3, 175)***

Note. *p < .05; **p < .01; ***p < .001.Columns related to step 1 and 2 indicate unstandardised Coefficients (B) with standard errors(SE), standardised coefficients (β), and t values. Gender is recoded as 0 = female and 1 = male. Results are based on 1000 bootstrap samples.