INTRODUCTION

Communication is essential for the proper interaction with another person or a group of people, as from the way we show our skills to express ourselves, difficulties may arise in different circumstances and be a constant. For that reason, communication skills are essential to socialize because they are within people's own abilities that are executed for communicative purposes and expressed in the behavior of different individuals. (^{O´Connor, 1999}).

Communication skills show a relationship with self-confidence and a solid behavior in the professional area, being fundamental, as they are part of the soft skills (^{Viktorovna, et al., 2021}).

Previous studies evidenced that a communication skills training program educated and trained university students, showing that it was an effective tool for improving communication skills in the short term (^{Hernández-Jorge & De la Rosa, 2018}). Likewise, these skills are relevant to academic aspects and their acquisition occurs especially when promoted through feedback, more accurate evaluative judgments and regularization of the learning process (^{Valencia-Naranjo & Robles-Bello, 2022}).

Studies on communication skills are a topic of interest in support careers, some oriented to the academic field, and an in-depth review on this can be seen in ^{Hernández-Jorge & De la Rosa (2018}).

The Communication Skills Questionnaire - HABICOM (^{Hernández-Jorge & De la Rosa, 2018}) provides an adequate measure of these skills. This questionnaire contains two subscales: the Personal Self-Perception (PA) scale and the Professional Importance (PI) scale. Each scale is composed of 34 items, with a Likert-type response format, using five levels for each one, being (1 = I do not use it and 5 = I use it a lot) for the first scale and (1 = it is not important and 5 = it is very important) for the second one. The PA scale contains eight factors: Generating Motivation, Nonverbal Communication, Empathy, Emotional Expression, Oral Expression, Informative Transmission, Open Communication and Listening and the PI Scale contains five factors: Emotional and Affective Quality, Informative Transmission, Nonverbal Communication, Open and Honest Communication and Openness and Authenticity.

Therefore, this study will analyze the psychometric properties and gender invariance of the HABICOM questionnaire so that it is reliable and valid, using a sample of Peruvian university students.

METHOD

Data Analysis

In order to perform an optimal analysis, the R statistic version 4.0.5 (^{R Core Team, 2020}) and its environment RStudio version 1.4.1106 (RStudio Team, 2020) were used. The Lavaan (^{Rosseel, 2012}) and semPlot (^{Epskamp et al., 2019}) packages were used to perform the Confirmatory Factor Analysis (CFA). The estimation method and the adjustment indices used were: Robust Weighted Least Squares (WLSMV), the Comparative Fit Index (CFI), the Root Mean Square Error of Approximation (RMSEA), Tucker-Lewis Index (TLI) and the Standardized Ratio Mean Residual (SRMR). And for the proper evaluation of these indexes, the following criteria were taken into account, for the CFI and TLI, values ≥ .90 and ≥ .95 were considered an adequate fit and good fit, values less than ≤ .08 and ≤ .05 in the RMSEA were also established as correct and finally for the SRMR, values ≤ .08 and ≤ .06 were considered correct as well (^{Keith, 2015}).

In the case of measurement invariance, the criteria for evaluating were set as follows: ΔCFI ≥ .010 and ΔRMSEA ≥.015 (^{Chen, 2007}; ^{Svetina et al., 2019}).

Table 1 Modified Items 

Source. Elaborated by the author.

RESULTS

Reliability Analysis

The internal consistency was adequate for the factors of the two subscales, with alphas ranging from .63 to .85 for the PS subscale and with alphas ranging from .75 to .91 for the PI. The information on the internal consistency of each factor can be seen in Table 3, Table 4 and Table 5.

Table 2 Descriptive statistics of the HABICOM items 

N = Total sample, M = Media, DS = Standard Deviation, Asi = Asymmetry, Cur = Kurtosis.

Source. Elaborated by the author.

Gender Invariance

To perform the invariance, the dimensionality of the questionnaire was evaluated to see if it was equivalent according to gender. Table 6 shows the results obtained, where the PS subscale reached the scalar level correctly (this level being required to compare scores between men and women), reaching the residual or strict level. However, in the PI subscale it was not possible to obtain optimal results, the modification indexes were consulted in order to make the necessary changes, but as can be seen in the ΔCFI metric level, this exceeds what was established.

Table 3 Standardized factor loadings of the PS Subscale 

Source. Elaborated by the author.

Table 4 Standardized factor loadings of the PI Subscale 

Source. Elaborated by the author.

Table 5 Results of the Confirmatory Factor Analysis 

χ2 = chi-square; df = degrees of freedom; RMSEA = root mean square error of approximation; CFI = comparative fit index; SRMR = Standardized root mean square residual; TLI = Tucker-Lewis Index

Source. Elaborated by the author.

Table 6 Gender Invariance Results 

M1 = Configural invariance; M2 = Metric invariance; M3 = Scalar invariance; M4 = Residual invariance; χ2 = chi-square; df = degrees of freedom; CFI = comparative fit index; RMSEA = root mean square error of approximation; ΔCFI = increase in the comparative fit index; ΔRMSEA = increase of the mean square error approximation.

Source. Elaborated by the author.

DISCUSSION

In psychology, interpersonal communication is relevant because it is inherent to the person. It is a resource that facilitates interaction with others, allowing an adequate social development.

For ^{Hernández-Jorge & De la Rosa (2018}), with respect to the communication skills variable, few studies referring to this variable in university students have been found. However, in the study conducted by (^{Tejera & Cardoso, 2015}) in Cuba mention that communication skills in the academic environment contribute to the achievement of better academic results, relationships with others (teachers, peers and family), in physical, mental, functional and subjective health as well as in their development in society. In the study conducted by ^{Hernández-Jorge & De la Rosa (2018}), regarding the use of communication skills, it was found that, university students obtained acceptable scores in most of the factors studied. This means then that communication skills are perceived by students as acceptable, being the empathic attitude the one that stands out over the expression of emotions which is found in a lower degree.

The HABICOM questionnaire has not been validated in the Peruvian population. Therefore, our study had two objectives: to evaluate the psychometric properties and gender invariance of the questionnaire. In general, the questionnaire showed adequate psychometric properties, since the indices obtained are within the established range, and reliability was confirmed.

No studies on communicative skills using the HABICOM have been found in our setting. That is why our research has been oriented to the analysis of the psychometric properties and invariance.

During the validation process, no significant problems were found, since consensus was satisfactorily reached during the review of the items by the experts. The CFA of each subscale showed satisfactory results, showing consistency with the multidimensional structure proposed in the original study (^{Hernández-Jorge y De la Rosa, 2018}).

Unlike ^{Valencia-Naranjo & Robles-Bello (2022}) who reduced through CFA the HABICOM scale to 26 items in 7 factors (eliminating the Informative Transmission factor), but only for the PS subscale and not for the other subscale, these results in general show that studies on the HABICOM should be analyzed with other samples, since both the results of Valencia-Naranjo & Robles-Bello (2022) and those of this study are favorable, with specific emphasis on the PI subscale, since what was shown in this study can make way for their own analyses.

Referring to the above, the invariance in the PS subscale did not show problems, unlike the PI subscale, since in the latter it was not possible to establish a satisfactory invariance, which calls for further research to explain it. Being the first time that the gender invariance of this questionnaire has been performed, it is necessary to carry out other studies in different populations and considering other sociodemographic variables in order to make a deeper analysis of the structure of this subscale.

With respect to reliability, adequate coefficients were obtained, as shown by the factors of the PS subscale, with values between α = .63 y α = .85, similar to the ^{Valencia-Naranjo & Robles-Bello study (2022}) α = .635 and α = .791. And for the PI subscale, the resulting alpha coefficients ranged between α = .75 and α = .91.

This study allows the use of the HABICOM questionnaire in Peruvian university students, considering that it shows reliable and valid measures; however, this research is not free of limitations, which should be taken into account for other studies: 1) Non-random sampling does not allow for replicability of results, 2) These results are considered acceptable, due to the relatively large sample size, however, this should be analyzed more rigorously with other samples and 3) The psychometric analysis performed could be considered as another limitation, since other evidence of validity was not considered, so this should be considered for future research, specifically the evidence of validity with other constructs.

In conclusion, this research shows that the HABICOM questionnaire is an adequate instrument to study communicative skills of university students in our environment. This questionnaire is a valid test with a multifactorial structure and adequate reliability indexes. Therefore, it constitutes a contribution to be used in empirical, instrumental or theoretical research