Chapter 4: Data Analysis and Results
4.6 Hypothesis
In the hypothesis testing, the statistical significance indicates whether the relationship between the variables occurs by chance, or whether it occurs due to the characteristics of the variables. The hypothesis testing is done at a confidence level of 95%, which implies that the results reveal that the relationship represented by the hypothesis is determined to occur in the manner predicted with a 95% confidence level.
In instances where the statistical significance does not meet the threshold, the relationship is indicated not to occur as stated.
4.6.1 Direct Effects
The following hypothesizes are to be tested for direct impact relationship:
H1: perceived organizational support has a direct and positive relationship with organizational identification during post-merger integration, i.e. β POS to OID = 0.
H2: perceived organizational support has a direct and positive relationship with organizational trust during post-merger integration, i.e. β POS to OT = 0.
H3: employees Participation in decision making has a direct and positive relationship with organizational identification during post-merger integration, i.e. β POS to OID
= 0.
H4: employees Participation in decision making has a direct and positive relationship with organizational during post-merger integration, i.e. β POS to OID = 0.
4.6.2 Mediation
The most common method to establish mediation is the causal model approach.
Such approach involves estimating each path in the model (Figure 7) and check
whether certain statistical criteria are met. However, one of the requirements of mediation is the statistical significance of the total effect. They claim that a non- significant total effect (path c) should not warrant further investigation of mediation.
Thus, new alternative approaches were suggested the most common of which is to use bootstrapping. The Bootstrapping method involves the use of bootstrapped samples to estimate the indirect effect (the ab path). For each sample, the estimate of path ab is calculated. The pooled estimates are then used to construct the bootstrapped 95%
confidence interval for the indirect effect. Confidence intervals that do not include 0 were considered statistically significant and the latter approach was used to test for mediation. Such approach does not require the total effect (c) to be statistically significant but only the indirect effect (ab). Noteworthy, 2000 bootstrapped samples will be used to test for the statistical significance of mediation.
Figure 7: Diagrammatic representation of mediation
Complete mediation is present when variable X no longer affects Y after M has been controlled, making path c' zero i.e. β = 0. Partial mediation is deemed present when the path from X to Y is reduced in absolute size but is still different from zero when the mediator is introduced. Both types require that the path ab is statistically significant i.e. 95% CI ≠ 0.
The following are the list of hypotheses that are to be tested for mediation as per the hypothesized theoretical model:
H5a: organizational identification mediates the relationship between perceived organizational support and employees change supportive behavior during post-merger integration, or OID mediates the relation between POS and ECSB i.e. β indirect = 0.
H5b: organizational identification mediates the relationship between employees participation in decision making and employees change supportive behavior during post-merger integration, or OID mediates the relation between EPDM and ECSB i.e.
β indirect= 0.
H6a: organizational trust mediates the relationship between perceived organizational support and employees change supportive behavior during post-merger integration, or OT mediates the relation between POS and ECSB i.e. β indirect= 0.
H6b: organizational trust mediates the relationship between perceived organizational support and employees change supportive behavior during post-merger integration, or OT mediates the relation between EPDM and ECSB i.e. β indirect= 0.
H7: Employees change supportive behavior mediates the relation between organizational identification and merger success during post-merger integration, or ECSM mediates the relation between OID and MS i.e. β indirect= 0.
H8: Employees change supportive behavior mediates the relation between organizational trust and merger success during post-merger integration, or ECSB mediates the relation between OT and MS i.e. β indirect= 0.
4.6.3 Moderation
Moderation analysis, also referred to as interaction analysis, is used to test whether the relationship between two variables is dependent on a third variable. In this Section, moderation is performed between a multiplicity of continuous independent variables and continuous dependent variables.
The following relationships are to be tested for moderation:
H7a: communication during post-merger integration moderates the relationship between perceived organizational support and organizational identification, or MCI1 moderates the association between POS and OID.
H7b: training during post-merger integration moderates the relationship between perceived organizational support and organizational identification, or MCI2 moderates the association between POS and OID.
H8a: communication during post-merger integration moderates the relationship between employees’ participation in decision making and organizational identification, or MCI1 moderates the association between EPDM and OID.
H8b: training during post-merger integration moderates the relationship between employees’ participation in decision making and organizational identification, or MCI2 moderates the association between EPDM and OID.
H9a: communication during post-merger integration moderates the relationship between perceived organizational support and organizational trust, or MCI1 moderates the association between POS and OT.
H9b: training during post-merger integration moderates the relationship between perceived organizational support and organizational trust, or MCI2 moderates the association between POS and OT.
H10a: communication during post-merger integration moderates the relationship between employees’ participation in decision making and organizational trust, or MCI1 moderates the association between EPDM and OT.
H10b: training during post-merger integration moderates the relationship between employees’ participation in decision making and organizational trust, or MCI2 moderates the association between EPDM and OT.
H11a: communication during post-merger integration moderates the relationship between organizational identification and employees change supportive behavior, or MCI1 moderates the association between OID and ECSB.
H11b: training during post-merger integration moderates the relationship between organizational identification and employees change supportive behavior, or MCI2 moderates the association between OID and ECSB.
H12a: communication during post-merger integration moderates the relationship between employees’ participation in decision making and employees change supportive behavior, or MCI1 moderates the association between OT and ECSB.
H12b: training during post-merger integration moderates the relationship between employees’ participation in decision making and employees change supportive behavior, or MCI2 moderates the association between OT and ECSB.