*4.3. Descriptive Analysis*

From the descriptive analysis of the items, the general conclusion is that there is a high appreciation of the "Function" related strategies since the average of responses to all items is 4.14 with an agreemen<sup>t</sup> of 86.92% (Table 4). These results make it possible to identify the EB strategy with a more positive perception on the part of employees. The less valued strategy by employees is "Management processes", with a mean of 3.13 and an agreemen<sup>t</sup> = 45.93%. These results allow us to identify it as the EB strategy with the least positive perception on the part of employees.


**Table 4.** Descriptive analysis.

#### *4.4. Correlation Analysis Parametric Test and Non-Parametric Test*

To further solidify this investigation, we applied Spearman's Correlation Test (Table 5). Some nonparametric tests can handle ordinal data, ranked data, and not be seriously a ffected by outliers. Non-parametric tests were applied to this investigation due to having Likert data and wanting to compare two or more groups. Spearman's correlation coe fficient is a statistical measure of the strength of a relationship between paired data. The strength of the correlation can be described as follows:

0.00–0.19 "very weak"; 0.20–0.39 "weak"; 0.40–0.59 "moderate"; 0.60–0.79 "strong"; and 0.80–1.0 "very strong".


According to the Spearman's rank correlation coe fficient, a correlation of 0.776 (Table 5) means that there is a *very* strong correlation between EB and AOC.


**Table 5.** Non-parametric test: Spearman's correlation coe fficient.

Notes: \*\* Correlation is significant at the 0.01 level (2-tailed).

#### *4.5. Measure of Sampling Adequacy*

The Kaiser-Meyer-Olkin Test (KMO) as a Measure of Sampling Adequacy was used in the present study (Table 6) to detect multicollinearity in the data to verify the appropriateness of carrying out a factor analysis. More specifically, sampling adequacy predicts if data are likely to factor well, based on correlations and partial correlations. A value above 0.8 is considered excellent. The Bartlett's Test of Sphericity is another test of the strength of the relationship among variables.


**Table 6.** Employer branding (EB) Kaiser-Meyer-Olkin (KMO) Bartlett test.

For our data, KMO was 0.913 for the EB scale and 0.833 (Table 6) for the AOC scale, signaling that a factor analysis of the variables can proceed. The results of our analysis for the Bartlett's Test of Sphericity showed a significance level of 0.00. It can be concluded that the relationship among variables is strong.

#### *4.6. Hypotheses Testing Framework*

As for the first group of hypotheses, a Spearman's correlational analysis was run to determine the relationship between each dimension of EB and AOC. It was found that all dimensions of EB have a positive correlation with the affective component of organizational commitment. It appears that the strategies "mission, vision, and values" respond to the specific objective: "Identify which employer branding strategy has the greatest positive impact on affective organizational commitment".

From the results obtained, we can also extract that the EB strategy with less positive impact on AOC corresponds to "compensation and benefits", thus responding to another of the specific objectives of the research.

Regarding the second group, the hypothesis "affective organizational commitment has a positive relation with employer branding strategies", it is confirmed that the affective component has a positive impact on the perception of EB strategies, by employees (*<sup>r</sup>*s = 0.180, n = 172, *p* = 0.001).

A third group of hypotheses was set to test the effect of sociodemographic variables on AOC. Spearman's correlational analysis was performed, as well as Mann–Whitney U non-parametric tests and the Wilcoxon Test, to test the effect of gender; and Kruskal–Wallis non-parametric test H, for the remaining sociodemographic variables. In relation to the hypothesis: "sociodemographic variables have a positive impact on employees' affective organizational commitment", we have that this hypothesis studies the relationship between sociodemographic variables and the affective component of organizational commitment. The results of the Spearman's correlational analysis indicate that the hypothesis was not validated (*p* > 0.001). Since there was no positive association between the sociodemographic variables, as a whole and the affective commitment, the impact of each sociodemographic variable on the affective component of the organizational commitment was tested individually, in the next step of the statistical treatment of the data, where it was demonstrated that the sociodemographic variables age group, gender, employment relationship, and sector of activity have a positive impact on the AOC of employees, but literacy and seniority do not.

In the fourth group, the hypothesis "The combined effect of sociodemographic variables with employer branding strategies on affective organizational commitment is positive" was tested by performing the chi-square test of independence (descriptive analysis with a cross-reference table). This was done to verify if there is a relationship between the variables EB and SDV. The results indicate that X<sup>2</sup> = 608.710 and *p* = 0.093; therefore, the association was not significant (*p* ≤ 0.05). The hypothesis was not confirmed. Therefore, EB and sociodemographic variables are not positively related to AOC. As previously verified, EB has a positive relationship with AOC; however, sociodemographic variables, as a whole, have no positive relationship with AOC. In summary, it appears that EB variables and sociodemographic variables do not act in a combined way on AOC, but independently. All the results are summarized in .

Table 7 presents the validated and not validated hypotheses. As for the sociodemographic variables, it was not possible to verify all the hypotheses, so it was necessary to reformulate this part of the conceptual model. The original conceptual model was redefined, by excluding paths of hypotheses that were not validated—H3, H3c, H3d, and H4. Once all the hypotheses regarding the relationship between EB dimensions and AOC have been proven, this attributes relevance to the conceptual model of this research study, due to the fact that all dimensions related to EB strategies proved to be relevant, even when tested individually.



This study had as its main objective to answer the following research question: "What is the impact of EB strategies, implemented by organizations, on AOC?", which was answered through the hypothesis test "H1—EB strategies have a positive impact on the employees' AOC". In order to deepen the analysis of this relationship, a Spearman's correlational analysis was carried out, the results of which confirmed the existence of a positive and significant correlation between perceived EB strategies

and the AOC of individuals. The constructs showed a strong and significant correlation at the 0.01 level, observing a correlation of 0.776 according to the Spearman table. Thus, results obtained with these techniques empirically support the following objective "to verify if there is a correlation between the EB constructs and AOC". Thus, we conclude that the perception of EB strategies is significantly correlated to the level of a ffective commitment of employees, realizing that the better the perception of the employee in relation to the strategies associated with EB, the greater his level of a ffective commitment, consequently raising your desire to stay in the organization. The various human resource managemen<sup>t</sup> strategies must be integrated into the EB process, covering all of the areas of the organization, namely: communication and image, operations management, marketing, culture, mission and values, training and development, and recruitment processes [46]. EB has been gaining importance due to the growing focus on factors, such as the power of branding, the commitment of employees, the war for talent, and the role of human resource managemen<sup>t</sup> in business [11].

A Spearman's correlational analysis was performed for each of the dimensions of the EB variable, noting the existence of a positive and significant correlation between each of the various dimensions in isolation with AOC, which allowed all of the hypotheses in the first group to be empirically supported. Thus, confirming the theoretical support for each one of the EB dimensions (function, compensation, mission, vision and values, team, career development, culture and work environment, managemen<sup>t</sup> processes, communication and image) as follows below.

The hypothesis "H1a—EB strategies related to the function have a positive impact on the employees' AOC": For Meyer and Allen [31], a ffective commitment develops when professionals' work experiences satisfy their needs for autonomy, skills development, and diversity and identification with tasks. Michaels et al. [2] state that talented professionals today want to feel motivated by the role they play, that there are enriching development and career opportunities, and that leaders support and motivate them. Organizations must be able to respond to this rise in expectations.

The hypothesis "H1b—Strategies related to compensation and incentives have a positive impact on the employees' AOC": Minchington [57] defines EVP (employee value proposition) as a package of attractive benefits that employers o ffer to employees, such as salary, bonuses, career development, work environment, possibility of promotion, and flexibility of working hours. In addition, EVP should not be focused only on monetary factors, such as compensation, but on psychological factors because they have greater weight among individuals [58].

The hypothesis "H1c—Strategies related to the mission, vision and values have a positive impact on the AOC of employees": Mowday et al. [59] present the main characteristics of a ffective commitment: the belief and acceptance of organizational values and objectives, the desire to maintain the link with the organization and a willingness to defend the organization. According to Fernandez-Lores et al. [46], when the relationship between the employee and the employer brand is established, the a ffective commitment expresses the attachment that the professional experiences towards the values of the employer's brand. And this attachment is manifested by the desire to remain in the organization over time and in the a ffinity of values between both parties.

The hypothesis "H1d—Strategies related to the team have a positive impact on the AOC of employees": Meyer and Allen [31] stated that the relationship between team behaviors and a ffective commitment would be di fficult to be studied, given that the results in their investigation were contradictory. This fact can be explained by the respondents' personal characteristics in view of the organization's relationships and internal environment.

The hypothesis "H1e—Strategies related to career development have a positive impact on the employees' AOC": According to Noe and Wilk [60], the training and skills development strategies implemented in career management, result in an increase in the level of AOC of their employees.

The hypothesis "H1f—The strategies related to culture and the work environment have a positive impact on the employees' AOC": For Miles and Mangold [25], the way and breadth with which the organizational culture is perceived and accepted by employees has an important impact on their behavior. The most important element of an organizational culture is its charter of principles and

values. Therefore, a correct managemen<sup>t</sup> of these values facilitates a ffective commitment since it leads to an approximation between the employee and the employer's brand—when he identifies himself with its values.

The hypothesis "H1g—Strategies related to managemen<sup>t</sup> processes have a positive impact on employees' AOC": According to Punjaisri and Wilson [51], the managemen<sup>t</sup> processes, from the initial moment of recruitment and selection, to the reception and integration phases, to the evaluation and managemen<sup>t</sup> of performance, all of these managemen<sup>t</sup> processes must be aligned with the brand values of the employer in order to increase the levels of a ffective commitment of individuals towards the employer. According to King et al. [61], it is possible to increase the levels of a ffective commitment of individuals towards the employer through leadership practices when they correctly represent the employer's brand values.

The hypothesis "H1h—Strategies related to communication and image have a positive impact on the employees' AOC": Results allowed us to conclude that the perception of EB strategies related to communication and image is significantly predictive of the level of a ffective commitment of employees. For King et al. [61], the values of the employer's brand must be e ffectively communicated so that they are assumed by employees, raising their levels of commitment to the employer's brand.

As for the specific objective, "EB strategies have a positive perception on the part of employees", it is possible to verify the results of this in the descriptive analysis of the items that make up the "QEB" scale. The average, as a measure of central tendency, allows knowing how the responses to the questionnaire are distributed, in the sense of knowing which EB strategies are most valued and least valued by the respondents. In this sense, it is possible to observe that all dimensions present M > 3.13. Since the Likert scale ranges from 1 to 5, this means that all responses are, on average, positive.

The specific objective "the EB strategy with a more positive perception on the part of employees" is verified by the descriptive analysis, where it is observed that there is a high valuation of the "Function" strategy, the same happening in the study by Sezões [14], since the average of responses to all items is M = 4.14 with an agreemen<sup>t</sup> of 86.92%. Likewise, the standard deviation of all items is less than 1, which reveals a high concentration around the average value = 5, confirming the homogeneity of the responses, at the highest values of the scale (between 4 "agree" and 5 "agree totally"). The median value (central value in each item) is always 4 or 5, which also reveals a high level of agreemen<sup>t</sup> among respondents. "Function" corresponds, therefore, to "the strategy of EB with a more positive perception on the part of employees".

As for the objective "to identify the EB strategy with a less positive perception on the part of the employees", it is observed that the strategy perceived in a less positive or that is less valued by the employees is "management processes", presenting an average value M = 3.13 and agreemen<sup>t</sup> = 45.93%. The "compensation and incentives" strategy following, with M = 3.37 and degree of agreemen<sup>t</sup> in the respondents' answers of 51.31%, the latter being the least valued strategy in the study by Sezões [14].

In relation to the specific objective "to verify if there is an association and positive and significant correlation between the dimensions that measure the EB strategies and AOC, the Spearman correlation coe fficient reveals that the items in the EB variable have a positive and significant correlation with all items in the AOC variable, this correlation being significant at the 0.01 level. All variables have a strong correlation (≥ 0.5). These results allow us to respond positively to the objective "to verify if there is a positive and significant association and correlation between the dimensions that measure EB strategies and AOC".

Regarding the hypothesis "H2—AOC has a positive impact on EB strategies", a Spearman's correlation was run to determine the relationship between AOC and EB values. Results confirm that this hypothesis is validated (*<sup>r</sup>*s = 0.180, n = 172, *p* = 0.001). According to Across [52], a greater perception of organizational commitment is associated with a greater perception of dimensions, such as: compensation, career, communication, supervision, and nature of the function.

To assess the specific objective "to verify if employees feel committed, at an a ffective level with the organization", according to the results of the descriptive analysis as for the variable AOC, M = 3.63 and

with an agreemen<sup>t</sup> greater than 40% in items 1, 2, and 5, respectively: "I would be very happy to spend the rest of my career in this organization"; "I feel the problems of this organization as if they were mine"; and "This organization has grea<sup>t</sup> personal significance for me". Interestingly, there is a high "disagreement", exceeding 36% in the questions formulated in the negative: "I don't feel like" being part of the family "in this organization"; "I don't feel" emotionally attached "to this organization." From the analysis of these values, it is possible to conclude that, on average, employees are emotionally committed to their organization.

Regarding "H3—SDV have a positive impact on employees'", a Spearman's correlation was run to determine the relationship between SDV and AOC values. Results confirm that this hypothesis is not validated (*<sup>r</sup>*s = 0.71, n = 172, *p* > 0.001).

In this case, we can conclude that the sociodemographic variables, considered as a whole, do not constitute a direct predictor of the level of AOC. Meyer et al. [53] claim that certain sociodemographic variables, such as age, seniority, or gender, influence AOC. Meyer and Allen [31] argue that personal characteristics can be considered as a factor that introduces uncertainty in the percentage of unknown variance, that is, in the variance of AOC that is not explained. Since there was no positive association between the sociodemographic variables, as a whole and the a ffective commitment, the impact of each sociodemographic variable on the a ffective component of the organizational commitment was tested individually, in the next step of the statistical treatment of the data.

H4 is not confirmed (X<sup>2</sup> = 608.710 and *p* = 0.093). As previously verified, EB has a positive relationship with AOC; however, SDV, as a whole, have no positive relationship with AOC. In summary, it appears that the variables EB and SDV do not act in a combined way on AOC, but independently. Additionally, there are divergent results when testing each SDV individually. Meyer et al. [53] claim that certain sociodemographic variables, such as age, seniority, or gender, influence AOC. In turn, for Rego and Souto [62], there is no significant correlation between the variable's seniority and educational qualifications for a ffective commitment, but it does exist for gender and age.
