Dataset on the safety behavior among Pakistani healthcare workers during COVID-19

The dataset includes the particulars of 515 respondents on safety behavior during COVID-19. The questionnaires were adapted using Social Learning Theory and Social Exchange Theory. The variables included in dataset are Transactional Leadership (TSL), Transformational Leadership (TFL), Employee Well-Being (EWB) and Safety behavior (SB). Moreover, the dataset also contains the demographic profile of the respondents. Data was collected with the help of self-administered questionnaire from eight public hospitals in Punjab, Pakistan, namely Services Hospital Lahore, Sir Ganga Ram Hospital Lahore, Government General Hospital Faisalabad, DHQ Hospital Chiniot, Municipal General Hospital Sargodha, DHQ Hospital Jhang, DHQ Hospital Multan and Sulehri Children & General Hospital Sialkot. This dataset could provide a significant insight for future research in employee safety behavior.


Specifications
Organizational behavior Specific subject area Human resource management Type of data Table  How data was acquired Data was collected through self-administered questionnaire using five point likert scale. The data also includes demographic features of respondents. Data format Raw, analyzed Description of data collection Data was collected via self-administered questionnaire in Punjab province. Total 550 questionnaires were distributed and after removing incomplete questionnaires only 515 were used in data analysis. It took 3 months for data collection. Data

Value of the Data
• The dataset provide the evidence that organizations should implement workplace policies and procedures that influence safety behavior. • The dataset is useful to address other related issues. For instance: safety climate, safety culture and safety citizenship behavior. • The dataset is helpful to predict the attitude and behavior of employees at early stage of employment. • The dataset focused on the hurdles which slow down the leadership practices in order to achieve a safe work environment. • The dataset advocates that management must invest to satisfy the psychological needs of the workers to shine their basic workplace skills. • The dataset could be used by other researchers in order to compare this data with other data obtain from related studies but different geographic regions.

Data Description
The dataset includes the questions related to four constructs: Transactional leadership (TSL), Transformational leadership (TFL), Employee well-being (EWB) and Safety behavior (SB). Definitions of all constructs and references relating to instrument are given in Table 1 .
The SPSS sheet along with questionnaire are given as a supplementary file. The questionnaires for TSL and TFL were adapted [3] to explain the relationship by incorporating social learning theory. Similarly, the questionnaires for SB and EWB are also adapted [5 , 7] . The relationship between EWB and SB were explained by social exchange theory. Five-point Likert scale ranging (1 = strongly dis-agree, 2 = dis-agree, 3 = neutral, 4 = agree, 5 = strongly agree) was incorporated which enhance the quality of responses and lower the tiredness of the respondents. TSL consist of five items, whereas TFL consist of six items. Items for TSL and TFL were adapted from previous study [3] . In addition, items for EWB and SB were adapted from different studies [5 , 7] with three and seven items respectively. Structural Equation Modeling technique using smart PLS 3.2.6 has been incorporated to explain the measurement and structural model.

Materials and Methods
Data was gathered through convenience sampling. It is a type of non-probability technique in which the data is collected from the people easily to approach. Data is collected from the public hospitals of Punjab Pakistan. Province Punjab is selected for data collection as it is the second most populous province and is known for quality hospitals. Data was entered and coded in SPSS software. All the preliminary tests were conducted on SPSS. For the main analysis Smart PLS 3.2.6 was used.

Loadings, composite reliability and average variance extracted
Individual item reliability of construct above 0.30 can be retained [8] . Similarly, items must be removed if the removal increase the value of average variance extracted (AVE) and composite reliability (CR). Therefore, the present dataset removed one item (SC2) as the removal increase the value of AVE and CR. In addition, the value of AVE is above than 0.5 which means convergent validity is established. Table 2 shows the adequate, individual item reliability, CR and AVE of the study's constructs.    Tables 3 , 4 and 5 were in acceptable range.

Ethics Statement
It is stated that the consent was taken from each individual who participated in this survey.

Declaration of Competing Interest
The authors declared that there is no conflict of interest either financial or personal among them.

Supplementary Materials
Supplementary material associated with this article can be found in the online version at doi: 10.1016/j.dib.2022.107831 .