AD ALTA
JOURNAL OF INTERDISCIPLINARY RESEARCH
categories of factors shaping the decision-making of investors.
Dunning introduced a well-known OLI paradigm and motives
which are essential in decision-making on investment, like
advantages resulting from ownership and ownership rights,
advantages resulting from information on human resources and
new information and specific advantages resulting from a
locality (Dunning, 1977, 1979, 2001). As far as Slovak authors
are concerned, we can mention J. Tancosova (2013, 2014), who
analyze determinants and location, and their significance in
relation to an access to FDI, and T. Dudas (2010), who deals
with the significance of workforce.
Among other things, foreign direct investment also supports the
development of national economies, increases employment in
the regions and has a positive impact on the trade balance.
Foreign investors are entering new markets because of sales
diversification. Similarly, due to a skilled and cheap labor force,
either because of increased turnover or the opportunities of new
markets. In terms of employment, FDI entering the host country
has a significant impact on maintaining or increasing domestic
employment, growth in labor skills and wage growth.
The wage level is considered by several authors to be one of the
most important factors influencing the decision to invest in a
large number of economic sectors in transition economies.
Dunning (1979) argues that labor costs are a significant variable
for foreign direct investment investors in 1970, and remains a
significant variable over the 1990s, along with the existence of a
skilled and skilled workforce. (Paul, et al., 2014)
Bobenič Hintošová, A., Bruothová, M., Kubíková, Z., &
Ručinský R. (2018) have identified the level of gross wages and
the share of labour force with achieved at least secondary
education, as the most significant determinants with the positive
effect on FDI inflows.
3 Methodology
The correlation is a statistical method that compares the
dependencies between the two variables X and Y. It represents a
value that examines whether it is just a random event, or that the
values compared are dependent and to what extent. The
correlation value is expressed numerically, ranging from 1,
positive correlation, to -1, negative correlation. Thus, a positive
correlation result is a direct dependence, while a negative
correlation result is a dependency of the indirect. Thus, values
ranging from -0.5 to +0.5 are defined as a weak linear
dependence. The closer the resulting value of the compared
phenomena to 0, the less dependency between the variables
compared to each other. Thus, the correlation coefficient
measures the magnitude and strength of the dependence between
the two interval variables.
Pearson’s correlation coefficient expresses the degree of
dependence between two variables. The reader is called
covariance. The calculation is feasible only for those variables
that are specified at certain intervals. If the correlation between
two variables changes as 0.1, this correlation is called trivial. If
the correlation has reached 0.1-0.3, we know that there is a small
correlation between the two variables, which also indicates a
small dependence. The correlation value between 0.3 and 0.5
expresses the mean dependence if the correlation exceeds 0.5,
then we are talking about a large dependency. The correlation of
0.7-0.9 represents a very high correlation and thus a high
dependence of one variable from the other variable. If the
correlation is greater than 0.9, we call it perfect dependence.
However, when calculating the correlation, we must also
consider whether we compare variables of a similar nature,
otherwise the correlation is irrelevant (Hindls, 2007). The
formula for calculating the Pearson correlation coefficient is as
follows:
Formula 1 Pearson correlation coefficient
The determination coefficient represents the increased value of
the Pearson correlation coefficient. The value of this coefficient
indicates in what size the variability of one variable determines
the variability of the other variable. It is expressed as a
percentage. For example, if r = 4, the coefficient of
determination will be r2 = 0.16; 0.16x100 = 16%. This means
that 16% of the variability of one and the other variable is
determined together.
The Spearman coefficient is used for ordinal or interval variables
that do not have a classical distribution. If we use interval
variables to calculate, they must first be converted to ordinal
variables. The value of the calculation results does not change
and is expressed between -1 and 1.
Formula to calculate the Spearman coefficient:
Formula 2 Spearman coefficient
In this paper we are using basic statistical data from Statistical
office of Slovak republic and Slovak National Bank from the
period of 2003-2017. With this dataset we analyse FDI and
economic indicators using the Pearson correlation coefficient,
which examines the dependence between selected economic
indicators of Slovak economy.
4 Empirical results and discussion
Using the Pearson correlation coefficient r we first determined
the intensity of the relationship respectively. the strength of
statistical dependence between FDI and variables -
unemployment rate, average nominal wage and GDP in
Slovakia. In the second step, we performed a regression analysis
using linear regression, which tested the quality of the
relationship between FDI and variables unemployment rate,
average nominal wage and GDP in Slovakia. As for the time
series of all 4 variables, we used the National Bank and the SO
SR data. The test period was 2003 to 2017, indicating that the
number of variables for statistical testing was 15.
The inflow of FDI generally contributes to the creation of new
jobs and thus to the problem of unemployment. Unemployment
in Slovakia started to decline in the same period when the first
major foreign investments started to emerge in Slovakia.
Figure 1 FDI inflow to Slovakia and unemployment rate in
Slovakia in 2003 – 2017
Source: Authors’ results
The results of the correlation analysis pointed to the fact that between
FDI and the unemployment rate there is only a weak negative
dependence (correlation coefficient r = - 0.177) with p-value at p = 0,
- 70 -