Pioneering work on firm level entrepreneurial behaviour was done by D. Miller [6] , who built on the work of past scholars ( Mintzberg [16] ; Khandwalla [17] ; Miller & Friesen [18] , Miller [6] ) described this behaviour as firm’s display of innovative, risk-taking and proactive approach in their strategic decisions. Thus the foundation of firm level entrepreneurship construct was laid, with these three dimensions, which as per him co-vary. Firms were categorized as entrepreneurial or conservative depending on whether they have high or low levels of all three dimensions. Most researchers in the past have uniformly adopted this gestalt approach to firm level entrepreneurial orientation (EO) Construct. Lumpkin & Dess [1] advocated competitive aggressiveness and autonomy as two key dimensions of EO Construct in addition to existing the three dimensions and further argued that these dimensions act independently rather than co-va- rying and have a unique and distinct relationship with firm performance. They argued that the existing uni-dimensional approach to EO construct can be one of the possible reasons behind the inconsistent and at times contradictory findings of the EO-Firm Growth relationship. Competitive Aggressiveness, captures Miller’s [6] approach of “beating competitors to the punch”, suggesting how firm responds to threats. While autonomy characterizes firm’s propensity to act independently through freedom to its workforce, as strong layers of bureaucracy and old established traditions in firms don’t contribute to new-entry.

The debate about the dimensionality issue of EO Construct has caused research to examine the dimensions of this construct both independently and collectively. Theoretically it is possible that all five dimensions are valuable for a firm but it is also possible that only a few of them are valuable. Miller [6] held that only when a firm regularly undertakes innovative endeavours, at considerable risk, ahead of its competition, should it be considered entrepreneurial. Majority of the researchers have held that entrepreneurial orientation construct is better explained as a uni-dimensional construct where the dimensions co-vary (Covin & Slevin [19] ; Miller [6] ; Wiklund [5] ). But firms may be entrepreneurial and engage in the process of new venture creation even when they exhibit certain level of imitativeness than innovation [20] ). They suggested that firms with entrepreneurial approach may gain more from imitation than from being innovative. Entrepreneurs tend to be moderate risk-takers [21] or have a desire to avoid risk [22] but may have high inclination towards innovation [23] . Based on these arguments Lumpkin & Dess [1] argued that an attempt to restrict firms’ entrepreneurial behavior to situations displaying high level of all EO dimensions falls short of explaining various other forms of entrepreneurship. There is emerging literature advocating that each dimension of EO make a unique and differential contribution to firms’ entrepreneurial behaviour. The empirical findings supporting the uniqueness of individual dimensions of EO have been growing [24] [25] [26] . There is a strong possibility that some dimensions might have carried the other dimensions along who may have limited or insignificant influence on firm growth or even a negative influence while viewing EO as a uni-dimensional construct. Miller [14] while reflecting about the emergence of firm level entrepreneurship (EO) in the last three decades stressed the need to reexamine the EO Construct, hinting the need to view multidimensional nature of EO Construct comprising five independent dimensions.

Researchers have suggested that introduction of third variable while analyzing a relationship between two variables provides a better picture of the actual relationship. EO is considered necessary but insufficient for small firms to grow as researchers have highlighted that a bi-variate examination may be insufficient to explain EO and firm growth relationship [27] [28] . Venkataraman & Grant [29] suggested that it is necessary in entrepreneurship research to identify critical contingency variables and establish their roles. The external environment firms’ face today is very complex and uncertain and this brings to fore both opportunities and threats [13] and small firms have to evaluate these opportunities and threats while taking key strategic decisions. Small firms that are new and do not possess the managerial skills to respond to environmental trends by adopting entrepreneurial strategies, find it difficult to grow. Their success depends, in large part, on their entrepreneurial strategies indicating their responses to the opportunities and threats posed by external environment [30] [31] .

Researchers in the past have focused on direct relationship between resources and firm growth while less attention has been paid to effective utilization of resources through firm’s strategic & entrepreneurial orientation [32] . Covin and Slevin [33] affirmed that entrepreneurial orientation requires resources as it affects firms’ entrepreneurial strategy, which in turn affects its growth. EO would be beneficial to firms in identifying opportunities before others and these firms can organize their resources to take advantage of these opportunities. Eisenhardt & Martin [34] mentioned that along with resources firms entrepreneurial and strategic posture is important as it facilitates the effective management of resources into value creation. Resources provide small firms the much required cushion to design and experiment with innovative products/services or processes, which fosters a culture of experimentation leading to better performance [8] . Wiklund and Shepherd [7] noted that the availability of resources allows firms to experiment with new pro-active and risky strategies that may not have been suitable in a resource-constrained environment.

The first dimension examined in this study is innovativeness; a strong focus on innovativeness helps in entering new markets, strengthens hold in existing ones and creates new possibilities [35] . Kreiser et al. [25] studied more than 1,000 small firms from six countries and found that innovativeness, pro-activeness has positive influence on sales growth. Yang [36] studied 406 small firms in Taiwan and found significant positive relationship between innovativeness, risk-taking and their growth. Becherer and Maurer’s [37] examined presidents of 215 small firms and found a strong relationship between pro-activeness and sales growth. Lumpkin and Dess [38] examined 124 executives from 94 firms and found that pro-activeness has positive relationship with all performance indicators. Organizations that do not take risks lose their competitive edge resulting in losing their market share and are not able to sustain their strong position [6] [33] . Khandwalla [17] noticed strong relationship between risk-taking and firm growth. Competitive aggressiveness is viewed as initiating actions and then forcing the competitors to respond to them. Morgan & Strong [11] found a negative relationship between competitive aggressiveness and small business growth. Burgelman [39] mentioned that the driver for firm level entrepreneurship lies in the independent and autonomous initiatives of enterprising individuals within an organization.

H1: There would be significant relationship between innovativeness and growth of small firms.

H2: There would be significant relationship between pro-activeness and growth of small firms.

H3: There would be significant relationship between risk-taking and growth of small firms.

H4: There would be significant relationship between competitive aggressiveness and growth of small firms.

H5: There would be significant relationship between firm autonomy and growth of small firms.

External environment brings an element of uncertainty to organizations and this poses numerous challenges for organizations. McGrath & MacMillan [40] mentioned that environmental uncertainty can benefit firms if they have an entrepreneurial approach. So firms can mitigate the influence of uncertain environment or enhance its performance by adopting an entrepreneurial posture [41] . Firms that focus on entrepreneurial strategy can maintain or even improve their performance in uncertain environment by exploring and exploiting new opportunities [42] . Zahra [8] found a strong relationship between entrepreneurship and growth among firms in uncertain environment. Specific aspects of entrepreneurial behaviour have been linked to uncertain environment. Miller and Friesen [18] posited that firms can adjust to environmental pressures through innovations in products or markets, use of proactive approach, by taking risk and competing aggressively. Khandwalla [17] pointed out that organizations respond to challenging environmental conditions by taking risk, which improves small firm growth. Miller and Friesen [18] found that pro-activeness was strongly associated with environmental uncertainty.

H6a: Environmental uncertainty would moderate the relationship between innovativeness and growth of small firms significantly.

H6b: Environmental uncertainty would moderate the relationship between pro-activeness and growth of small firms significantly.

H6c: Environmental uncertainty would moderate the relationship between risk taking and growth of small firms significantly.

H6d: Environmental uncertainty would moderate the relationship between competitive aggressiveness and growth of small firms significantly.

H6e: Environmental uncertainty would moderate the relationship between autonomy and growth of small firms significantly.

Lumpkin and Dess [1] argued that small firms endowed with rich resources would be in a better position to follow entrepreneurial strategies. There is an increasing consensus that resources encourage experimentation such as introducing new products/services, entering new markets [43] . Resources facilitate the discovery of new entrepreneurial opportunities in a proactive manner by encouraging firm employees to envision how existing resources can be redeployed in new value creating activities by granting them the necessary autonomy. Resources shift the focus away from analyzing what is currently being done to what is possible to be done and subsequently “opens an opportunity for entrepreneurial discovery and exploitation” [44] . While resources create the base from which a firm competitive capacity departs, strategic and entrepreneurial orientation (EO) concerns the mean of reaching that goal. Thus, availability of resources would be advantageous as risk associated with entering new market would be substantially reduced or firms would be in a position to absorb the consequences of risky decisions or the other way round the firms are better prepared to fight competition fiercely with the availability of resources. Wiklund [5] found that knowledge resources moderate the relationship between EO and firm growth; similarly Wiklund & Shepherd [7] noticed that financial resources moderate the relationship between EO and small firm growth. Thus:

H7a: Resources would moderate the relationship between innovativeness and growth of small firms significantly.

H7b: Resources would moderate the relationship between pro-activeness and growth of small firms significantly.

H7c: Resources would moderate the relationship between risk-taking and growth of small firms significantly.

H7d: Resources would moderate the relationship between competitive aggressiveness and growth of small firms significantly.

H7e: Resources would moderate the relationship between autonomy and growth of small firms significantly.

Configurational approach suggests that firms when configured on various constructs do better than firms that align on two constructs or study the main-effect model. Researchers have stressed that using configuration approach to study various multivariate combinations can explain the complex interrelations between various moderating variables with far better predictive power than bivariate contingencies. Wiklund & Shepherd [7] found that firms perform better by aligning EO with financial resources and environment. Miller [45] examined the interaction among multiple variables using configuration approach and found that firms perform better when they align their entrepreneurial strategy with external environment and organizational structure. The opportunities of firm growth vary in uncertain environment and firms may have to pick specific aspects of entrepreneurial strategy that may result in firm growth further most entrepreneurial strategies are resource intensive. Gupta & Pandit [46] , examined the influence of innovation configured with environmental dynamism (exogenous) and resources (endogenous) for small and medium enterprises (SMEs) in India.

H8a: Small firm growth is better explained by configurations of innovativeness, access to resources in uncertain environment.

H8b: Small firm growth is better explained by configurations of pro-ac- tiveness, access to resources in uncertain environment.

H8c: Small firm growth is better explained by configurations of risk- taking, access to resources in uncertain environment.

H8d: Small firm growth is better explained by configurations of competitive aggressiveness, access to resources in uncertain environment.

H8e: Small firm growth is better explained by configurations of autonomy, access to resources in uncertain environment.

This research was field study and used cross-sectional research design through survey approach. The sampling frame was the database published by Small Industries Research Institute (SIRI), Delhi. From the sampling frame, a random sample of 2200 small firms spread across Delhi-NCR was shortlisted for data collection purposes. Prior appointments were taken through phone/mail and out of a total of 2200 small firms approached for personal appointments in Delhi- NCR, a total of 270 firms responded with appointments. Data was collected from 242 respondents representing 242 firms as single respondent was selected from each firm. Out of 242 responses received, 19 questionnaires were rejected because of incomplete information, so a final data of 223 firms was considered for data analysis. Out of total 223 firms, 115 operated in manufacturing sector and 108 operated in services sector and 167 of these were small firms and 56 were medium sized firms. Regarding the age of firm, 128 firms were set up before year 2005 and 95 firms were set up after year 2005, whereas size of firm was measured by number of full-time employees, 137 firms had less than 50 employees and 86 had more than 50 employees.

Reliability of measurement scales was tested using Cronbach alpha, which is useful in investigating reliability of multi-item interval level scales. Nunnally, [51] ; Hair et al. [52] , held that Cronbach alpha value of above 0.70 is considered reliable. Table 1 shows the results of reliability analysis of the scales used and their means and standard deviation. While examining Cronbach Alphas, it was

N = 223.

found that all the variables examined in this study have reliability value above 0.70, which is the threshold value [51] .

To examine whether the EO construct represent these five as independent dimensions an exploratory factor analysis using principal component method with varimax rotation was conducted on the 20 items used to measure the five dimensions of EO Construct. Before conducting factor analysis, Kaiser-Meyer- Olkin (KMO) was done to check factorability and sample adequacy and results found the value of KMO was .811, which is higher than 0.50, indicating that the data is very reliable and suitable. Further, Bartlett’s test of Sphericity for testing the significance was found to be significant, as indicated by the p-value (observed level of significance) corresponding to the chi-square statistic. Tabachnick and Fidell [53] mentioned that the choice of cutoff value of loadings is the preference of the researcher. For this research, factor loading above 0.522 was considered significant. While examining the results of factor analysis, all the items were found to have values above 0.522, so all the twenty items to measure the five dimensions were retained.

Table 2 shows the results of factor analysis. All five dimensions combined together explained 67.6% of variance. The data used in this research was collected through questionnaire and contain self-reported measures, this raises a concern about Common Method Bias, which can result in inflated or deflated observed relationships between the constructs. To overcome this problem, Harman One Factor Test was done as mentioned by Podsak off & Organ [54] to avoid the potential threat to validity. The variables were factor analyzed and results of un-rotated factor analysis were examined, which yielded five factors with Eigen values higher than one with no particular variable explaining substantial variance suggesting that common method bias was not a problem.

Results of correlation analysis i.e. Pearson Product Moment Correlation Coefficient, measures the magnitude and direction of linear relationships among the

variables. A careful examination of the correlation matrix indicated that though many variables were correlated but there was hardly any degree of overlap among the independent variables indicating no issue of multicollinearity, so all the independent variables examined were retained. While analyzing the correlation matrix, it was found that correlations among the variables were well below 0.70, and the highest degree of correlation between two independent variables was 0.541 between innovativeness and pro-activeness. Sales growth was significantly correlated with all five EO Dimensions except autonomy. Though most of the EO dimensions were positively related with sales growth but competitive aggressiveness was negatively related with sales growth. Similarly employment growth was significantly correlated with some EO Dimensions. Sales and employment growth were significantly correlated with each other and the degree of correlation was 0.581. The results of Correlation matrix provides a strong indication that exists distinct relationship between individual dimensions of EO and firm growth emphasizing that dimensions of EO Construct don’t co-vary. These findings signal that for small firm growth all the dimensions of EO are not equally important, similarly their degree of association varies as the growth parameter changes i.e. their degree of association with sales and employment growth is different (Table 3).

Hierarchical linear regression analysis was used to test all the hypotheses as this is more appropriate. At each stage of the hierarchical regression, the next order of interaction is introduced (Universal Model, two-way and three-way interaction respectively) and incremental R² and F Tests of statistical significance were examined. The results are shown in four models; first the five independent variables were entered along with the main effect of the moderating variables namely environmental uncertainty and resources in the universal Model 1. Then the interaction terms were entered and the results of two-way interaction involving environmental uncertainty are shown in Contingency Model 2 and resources are shown in Contingency Model 3, lastly the outcome of three-way interaction employing configuration approach are displayed in Model 4. Tests for multicollinearity were done by computing tolerance values and variance inflation factor (VIF). These values ranged between 0.63 and 0.87 and the VIF values were less than 1.47, suggesting that multicollinearity was not the issue. In addition mean centering was done to overcome multicollinearity while testing the moderating variable, as the interaction term normally has higher probability of multicollinearity. The results of are shown in Table 4 and Table 5 respectively.

In Table 4 where the dependent variable is sales growth the results of Model 1 indicate that out of the five independent variables three were significantly related

**Correlation is significant at the 0.01 level (2-tailed); *Correlation is significant at the 0.05 level (2-tailed).

Standardized Regression Coefficients are displayed in the Table; *p < 0.05, **p < 0.01.

with sales growth except competitive aggressiveness and autonomy, the two variables of innovativeness and pro-activeness were significantly and positively related with sales growth whereas risk-taking was significantly but negatively related with sales growth as indicated by the beta values and p-values. The five independent variables and the main effect of environmental uncertainty and resources accounted for a variance of 28.9% in sales growth as indicated by ΔR². So the results support the first three hypotheses but reject the fourth and fifth hypothesis.

Standardized Regression Coefficients are displayed in the Table. *p < 0.05, **p < 0.01.

While examining the moderating influence of environmental uncertainty on the relationship between each of the five independent variables with sales growth, it was clear that the moderating influence varies with individual EO dimensions as indicated in Model 2. In case of sales growth, the moderating influence of environmental uncertainty was significant with innovativeness and risk- taking as indicated by the beta value and p-value while the moderating influence was positive with risk-taking it was negative with innovativeness whereas the influence was insignificant with the other three variables namely with pro- activeness, competitive aggressiveness and autonomy. The moderating influence of environmental uncertainty explained an additional variance of 2.0% in sales growth as indicated by ΔR². The findings support hypotheses 6a & 6c and reject hypotheses 6b, 6d & 6e.

While examining the moderating influence of resources on the relationship between each of the five independent variables with sales growth, it was clear that the moderating influence varies with individual EO dimensions as indicated in Model 3. In case of sales growth, the moderating influence of resources was significant with innovativeness and pro-activeness as indicated by the beta value and p-value whereas the influence was insignificant with the other three variables namely with risk-taking, competitive aggressiveness and autonomy. The moderating influence of resources explained an additional variance of 6.2% in sales growth as indicated by ΔR². The findings support hypotheses 7a & 7b and reject hypotheses 7c, 7d & 7e.

To examine the moderating influence of environmental uncertainty and resources on the relationship between each of the five independent variables with sales growth simultaneously using configuration approach, a three-way interaction was studied in Model 4. The results of three-way interaction were significant with four of the independent variables namely innovativeness, pro-active- ness, risk-taking and competitive aggressiveness as indicated by the beta value and p-value whereas the influence was insignificant with autonomy. This three- way interaction using configuration approach explained an additional variance of 19.6% in sales growth as indicated by ΔR². The findings support hypotheses 8a, 8b, 8c & 8d and reject hypotheses 8e. The findings of Model 4 clearly illustrate that the influence of three-way interaction using configuration approach was much stronger.

In the next step, the same process was repeated with employment growth as dependent variable in place of sales growth and the results are presented in Table 5.

In Table 5 where the dependent variable is employment growth the results of Model 1 indicate that out of the five independent variables two variables namely innovation and pro-activeness were significantly related with employment growth and the other three independent variables namely risk-taking, competitive aggressiveness and autonomy were not significantly related with employment growth as indicated by the beta values and p-values. The five independent variables and the main effect of environmental uncertainty and resources accounted for a variance of 10.6% in employment growth as indicated by ΔR². So the results support the first two hypotheses and don’t support the next three hypotheses.

Model 2 indicates that the moderating influence of environmental uncertainty on the relationship between each of the five dimensions of EO as independent variables and employment growth, was limited. The moderating influence was sig- nificant only with competitive aggressiveness while for the other four independent variables it was insignificant, as indicated by the beta and p-values. The moderating influence of explained an additional variance of 1.2% in employment growth as indicated by ΔR². Thus the findings support hypotheses 6d and reject hypotheses 6a, 6b, 6c & 6e.

The moderating influence of resources on the relationship between each of the five dimensions of EO and employment growth varied with individual dimensions as shown in the table. The moderating influence was significant with innovativeness and competitive aggressiveness whereas the influence was insignificant with the other three dimensions, as indicated by the beta and p-values. The moderating influence of resources explained an additional variance of 4.0% in employment growth as indicated by ΔR². Thus the findings support hypotheses 7a & 7d and reject hypotheses 7b, 7c & 7e.

To examine the simultaneous influence of environmental uncertainty and resources on the relationship between each of the five independent variables and employment growth using configuration approach, a three-way interaction was done as shown in Model 4. The results of three-way interaction were significant with three of the independent variables namely innovativeness, pro-activeness and competitive aggressiveness as indicated by the beta and p-values, whereas the influence was insignificant with risk-taking and autonomy. The three-way interaction model using configuration approach explained an additional variance of 10.8% in employment growth as indicated by ΔR². The findings support hypotheses 8a, 8b & 8d and reject hypotheses 8c & 8e. The results clearly illustrate that examining the EO dimensions and moderating variables using configuration approach show better results.