Results of the binomial logistic regression models
| Odds ratio | ||||
|---|---|---|---|---|
| Explanatory variables | Model 1 | Model 2 | Model 3 | Model 4 |
| (N = 347) | (N = 414) | (N = 198) | (N = 229) | |
| Providing online services before the COVID-19 pandemic | 0.694 | – | – | – |
| Possibility to convert a business model to an e-business model | – | 2.109* | – | – |
| Being impacted by COVID-19 (low impact) | – | – | – | 0.946 |
| Being impacted by COVID-19 (low impact) | – | – | 0.996 | – |
| Company size | ||||
| Small | 0.482 | 0.495 | 0.389** | 0.781 |
| Medium | 1.444 | 1.535 | 0.215 | 0.954 |
| Economy sector | ||||
| Primary | 0.310 | 0.344 | 0.594 | 0.594 |
| Secondary | 1.110 | 1.067 | 1.024 | 1.501 |
| Odds ratio | ||||
|---|---|---|---|---|
| Explanatory variables | Model 1 | Model 2 | Model 3 | Model 4 |
| ( | ( | ( | ( | |
| Providing online services before the COVID-19 pandemic | 0.694 | – | – | – |
| Possibility to convert a business model to an e-business model | – | 2.109 | – | – |
| Being impacted by COVID-19 (low impact) | – | – | – | 0.946 |
| Being impacted by COVID-19 (low impact) | – | – | 0.996 | – |
| Small | 0.482 | 0.495 | 0.389 | 0.781 |
| Medium | 1.444 | 1.535 | 0.215 | 0.954 |
| Primary | 0.310 | 0.344 | 0.594 | 0.594 |
| Secondary | 1.110 | 1.067 | 1.024 | 1.501 |
Significant at ***p < 0.001; **p < 0.01; and *p < 0.05. A “very small” category is a reference for a company size variable, and “tertiary” category is a reference for an economy sector variable. If a variable was not included in the model, then it is indicated by the “–” sign. The dependent variable for Models 1 and 2 is whether a company was impacted by COVID-19 or not (0-yes, 1-no); dependent variables in Models 3 and 4 present expectations of the companies about the impact of digitalization on “Digitization of processes” and “Innovative products and services” correspondingly (1-average to high impact, 0-low/very low impact). In Models 3 and 4, the independent variable; “Being impacted by COVID-19” is a dummy variable with categories of 1-low impact and 0-average to high impact