Skip to main content

Table 12 Regression results from an ordered probit model identifying the factors that influence the number of borrowing sources

From: Financial decision-making behaviors of Ethnic Tibetan Households based on mental accounting

Variable

Coeff

Marginal effects

0

1

2

3

4

Agrilost

0.145*

(0.086)

− 0.044*

(0.027)

− 0.003

(0.005)

0.031

(0.020)

0.014*

(0.008)

0.001

(0.001)

Labor

− 0.253***

(0.082)

0.077***

(0.026)

0.004

(0.009)

− 0.055***

(0.017)

− 0.025*

(0.014)

− 0.002

(0.002)

Dowry

0.027**

(0.012)

0.008**

(0.004)

0.000

(0.001)

− 0.006**

(0.003)

− 0.003*

(0.002)

− 0.000

(0.000)

loan info

− 0.354**

(0.148)

0.108**

(0.049)

0.006

(0.012)

− 0.077**

(0.033)

− 0.034*

(0.020)

− 0.003

(0.003)

Telecom

0.970***

(0.340)

− 0.296***

(0.104)

− 0.017

(0.037)

0.210***

(0.081)

0.094**

(0.048)

0.008

(0.008)

Health

− 0.183**

(0.076)

0.056**

(0.024)

0.003

(0.007)

− 0.040**

(0.019)

− 0.018**

(0.008)

− 0.001

(0.002)

  1. Coefficient not statistically significant and removed from the model. Ordered probit estimates (log-likelihood = -284.05, Wald \({\chi }^{2}\) = 64.92, and prob. > \({\chi }^{2}\)  = 0.00). The dependent variable is the number of sources Tibetan households borrowed from (ranging from 0 to 4). It's worth noting that among the independent variables, loan info refers to whether the respondent has ever heard about microfinance (yes = 1, no = 0). Standard errors in parentheses are obtained by village clustering
  2. *, **, and *** represent significant at 10%, 5%, and 1% level, respectively