Theoritical framework
Based on a literature review as well as from previous
studies, the variables involved in this study can be
formulated into the framework shown in Figure 1.
METHODOLOGY
Population and sample
The population used in this study is a manufacturing company that
is listed on the Indonesia Stock Exchange 2010 to 2012. This is
done for reasons that manufacturing companies typically do a lot of
CSR activities and their effect on the surrounding environment is
greater than other types of companies. The method used in
sampling is purposive sampling method, the sampling method is
based on certain criteria. 38 companies were selected.
Variables and measurement of variables
The variables consist of the independent variable, dependent and
intervening. Independent variables are variables that affect other
variables. The dependent variable is the variable that is affected by
the independent variable. Intervening variable is the variable or
variables that strengthen mediate between dependent and
independent variables.
In this study, earnings management is used as the independent
variable, the value of the company is used as the dependent
variable, and the disclosure of CSR used as an intervening
variable.
Earnings management as an independent variable in this study
was detected using a modified model of Jones (1991) discretionary
accruals proxy (discretionary current accruals). Model modified
Jones (1991) was used in this study because it is considered the
most excellent models in detecting earnings management. In
search of discretionary accruals, one must go through the following
steps:
Measure of Total Accrual (TACCit):
TACCit = NIit - CFOit
Measure of Non Discretionary Accrual (NDAit):
NDACCit = α1(1/TAit-1) + α2((ΔREVit-ΔRECit )/TAit-1) + α3(PPEit/TAit-1)
Measure of Disretionary Accrual (DAit)
DACCit = (TACCit / TAit-1) - NDACCit
CSR in this research are used as intervening variables. In this
study, the items used to measure the disclosure of social disclosure
is based on the ISO 26000 Guidance Standard on Social
Responsibility which consists of 33 items in 7 key themes such as
disclosures made by Handayati (2011). In this study, the measuring
instrument used to measure corporate social disclosure is
Dislousure Corporate Social Responsibility Index (CSRDI) based on
the items contained in the ISO 26000 Guidance Standard on Social
Responsibility. CSRDI formulated as follows:
CSRDI = (Items Numbers Disclosure/Total itmes number have to
disclosure)/100%
Firm’s value is used as the dependent variable. The firm’s value in
this study is defined as the market value. The firm’s value is
calculated by using the model of Tobin's Q Ratio developed by
Tobin (1967), which shows the estimated financial markets today.
When the Q ratio is above one, this indicates that the investment
will generate profits in assets that provide higher value than
investment spending, it will attract new investment. If the Q ratio is
below one, the investment is not interesting (Herawaty, 2008). The
calculation of the firm’s value is done using the following formula:
Q = (EMV + D)/(EBV + D)
Analysis technique
Regression models used to demonstrate the intervening variable
were calculated by creating two structural equation regression
equation, which suggests the hypothesized relationship. In this
case, there are two such equations, namely:
Firm’s value = α + p1 ML + p3 CSRDI + e2 (1)
CSRDI = α + p2 ML + e1 (2)
Path analysis
To test the effect of intervening variables, path analysis method
was used. Path analysis is an extension of the multiple linear
regression analysis or the use of regression analysis to assess the
causality relationship between variables (casual models) that have
been set previously. Path analysis alone cannot determine the
causal relationship and cannot be used as a substitute for the
researchers to see the relationship of causality between variables.
Causality relationship between variables has been established
based on theoretical models. W hat can be done by path analysis is
to determine the pattern of relationships of three or more variables
and it cannot be used to confirm or reject the hypothesis of
causality imaginary (Ghozali, 2011).
Path model proposed relationship is based on the theory that
earnings management has a direct relationship with the Company
Value (p1). Nevertheless earning management also has the effect
of indirect relationship to the firm’s value to disclosure of CSR (p2)
and then to firm’s value (p3). The total effect relationship of
earnings management to firm’s value (correlation between earnings
management and firm’s value) is equal to the direct influence of
earnings management to firm’s value (path or regression
coefficients p1) plus the indirect effect that the path coefficients of
earnings management to CSR disclosure is p2 multiplied by the
coefficient of the path of CSR disclosure to firm’s value is p3
(Ghozali, 2011).
The direct effect to the earning management to firm’s value = p1.
Indirect influence disclosure of CSR to earnings management = p2
× p3. The total effect (the correlation firm’s value to earnings
management = p1 + (p2 × p3)