Page 101 - 33-3

NTU Management Review Vol. 33 No. 3 Dec. 2023

ADV = β + β OFFICE_SIZE +β MGT_TENURE + β TARGET + β PRICE
i,t
0
i,t
4
1
i,t
3
i,t
i,t
2
+ β NUM_ENTRUST + β NUM_BRANCH +∑MONTH + ∑REGION +
i,t
i,t
6
5
ζ . (2)
i,t
3.5 Regression Models for Hypothesis 1
We used Ordinary Least Squares regression (OLS) models to determine whether the
level of goal consensus between the regional and branch managers is related to a branch
receiving a greater unexpected portion of senior salespersons and advertising funding.
We use the residuals from models (1) and (2) as the variables of interest and examine the
association between CONSENSUS and Ab_SENIOR and Ab_ADV .
i,t
i,t
i,t
Personal characteristics, environmental factors, and organizational structure
determine the decision-making process (Dean and Sharfman, 1996; Wally and Baum,
1994; Paolillo and Vitell, 2002; Sayegh, Anthony, and Perrewé, 2004). Following prior
studies, we include the regional manager’s age (REGIONAL_AGE ), gender (REGIONAL_
i,t
GENDER ), and tenure (REGIONAL_TENURE ) to control for the influence of these
i,t
i,t
characteristics on resource allocation decisions. We also include the branch office’s
ratio of senior salespersons in the previous month (allocated advertising funding in
previous month) to control for the lagged effect of distributed resources on the regional

manager’s resource allocation decision. In addition, we include the branch manager’s
age (MGT_AGE ), gender (MGT_GENDER ), and tenure (MGT_TENURE ) to control
i,t
i,t
i,t
for the effect of the branch manager’s characteristics on the regional manager’s resource
allocation decisions. We also include the average property price per deal sold by the
branch office (PRICE ), the number of houses managed by the branch office (NUM_
i,t
ENTRUST ), the office size (OFFICE_SIZE ), the office age (OFFICE_AGE ), and the
i,t
i,t
i,t
turnover rate (TURN ) to control for the impact of office characteristics on the regional
i,t
manager’s resource allocation decisions. Finally, we include the number of branch offices
in the same region (NUM_BRANCH ), the property transactions that occurred in the
i,t
administrative district (CYCLE ), the residential housing price index in the administrative
i,t
district (INDEX ), the month (MONTH), and the region effect (REGION) to control for
i,t
the influence of time and the environment on the regional manager’s resource allocation

decisions. Detailed variable definitions are presented in Table 2. Our models for H1 take
the following forms:

93