*4.1. Income Change Following a Move*

When comparing faculty members with a previous record of mobility with those who were never mobile, direct income was relatively similar; however, the indirect income of those who were immobile was much lower than that of the mobile faculty members. The results reveal that marked income changes occur for faculty members following a move, but unlike mobility in the industry sector, academic mobility leads to a decrease in faculty members' direct income. The findings presented in Table 2 indicate that for faculty members with mobility experience, a significant change in income occurred after their most recent move, with indirect income increasing and direct income decreasing.

Direct income. According to the survey responses, the total amount of the four direct income sources declined from ¥170,000 (approximately 26,282 US dollars) to ¥133,000 (\$20,563). Overall, after a move, the salary received from the respondent's HEI decreased by an average of ¥36,000 (\$5566), followed by decreases in income from other employers (down by ¥9000 or \$1391) and other academic institutions (down by ¥1000 or \$155). Only one item rose sharply: other income earned from the HEI (increased by ¥9000 or \$1391). The decrease in salary is due to the scholars' return to China from the U.S. or from European countries where their incomes were higher. On average, the incomes of Chinese scholars who returned from the U.S. decreased by ¥161,923 (\$25,033).


**2.**PercentchangeinincomeindicatorsresultingfromthemobilityoffacultymembersatChineseresearchuniversities.

\* BM = Before move; AM = After move; NM = No move.

Indirect income. Regarding the most recent move of faculty members with mobility experience, the amount of research funds received exhibited a dramatic 50% increase, from an annual average salary of ¥183,000 (\$28,293) to ¥274,000 (\$42,362). Vertical research funds primarily contributed to this increase. Vertical funds increased by 108%, while lateral funds increased by only 12%. In addition, the respondents reported a 72% increase in their spouses' total income, from an annual average of ¥85,000 (\$13,141) to ¥146,000 (\$22,572), and housing size also increased (although the increase was only approximately 1.6 square meters).

#### *4.2. Relativity Analysis of Income and Faculty Mobility*

As noted in Table 2, changes related to mobility were observed for both direct and indirect income. Does this mean that a relationship exists between income and faculty mobility? Further analysis suggested that direct income was positively associated (*r* = 0.198, *p* = 0.023) with the moving or not variable, and indirect income via total research funds had a significant positive correlation (*r* = 0.223, *p* = 0.004) with moving or not. Neither university geographic location (*r* = 0.098, *p* = 0.407) nor spouse's total income (*r* = 0.198, *p* = 0.023) was significantly associated with mobility. While spouse's total income was positively correlated with moving for female respondents (*r* = 0.086, *p* = 0.269), a similar correlation for male faculty (*r* = 0.287, *p* = 0.015) with mobility experience was not significant. This finding indicates that gender plays a special role in the relationship between income and mobility. The most likely reason is that a male's income is typically higher than his spouse's income. Spouses continue to maintain their salary levels following a move, a result that is consistent with some research on gender, family and faculty mobility [25,53].

Additional analyses revealed that mobility was positively correlated with indirect income. Additionally, mobility frequency was positively correlated with two of the indirect income indicators (*r* = 0.372, *p* < 0.001)—total research funds (*r* = 0.536, *p* = 0.002) and spouse's total income (*r* = 0.278, *p* = 0.002)—but was not significantly correlated with the university's geographic location (*p* = 0.220). These findings indicate that the mobility of faculty members at Chinese research universities significantly increases their indirect income, which may explain why some of these faculty members continue to pursue academic mobility regardless of decreases in their direct income.

Subsequent analyses employed two demographic indicators, gender and age, as control variables to perform partial correlation analysis. As noted in Table 3, gender did not make a difference in the relativity level between income (direct and indirect) and both moving or not and mobility frequency. However, when age was introduced, the correlation level between income (direct and indirect income) and moving or not disappeared. A significant positive correlation was observed only between current direct income and mobility frequency. Additionally, the correlation between mobility frequency and spouse's total income became non-significant when age was considered. This may indicate that when excluding the influence of age, a linear relationship between mobility frequency and direct income can be observed. However, when age is considered, the linear relationship between mobility and income disappears, and there is no relationship between mobility frequency and indirect income.


**Table 3.** Partial correlation matrix.

*Levels:* \*\*\* *p* < 0.001, \*\* *p* < 0.01, \* *p* < 0.05.
