Record Documentation Practice and Performance of Human Resouce Function at Moi Teaching and Referral Hospital, Eldoret Kenya

There was an increase in poor performance of human resource functions which led to negative consequences for an organization. Human resource departments were responsible for overseeing the administrative functions of an organization, including payroll, benefit administration, recruitment, and employee relations. Poor human resource management could lead to employee conflict, lack of recognition, inadequate training, and poor team building, among other issues. These issues could result in employee dissatisfaction, high turnover rates, and low morale. Therefore, the objective of this study was to investigate the influence of HR record documentation practice on the performance of the human resource function at Moi Teaching and Referral Hospital, Eldoret, Kenya. The study was guided by The Life cycle Model. The target population entailed employees of MTRH. 1 human resources manager, 3 benefits administrators, 18 health administrative officers, 86 training and development manager, 6 recruiter, 1 employee relations specialist and 65 HRIS Specialist. Therefore, the total accessible population was 180 respondents. The sample size of 124 respondents were drawn using the Yamane formula. The sample size was selected using the stratified sampling technique, with the strata representing the cadre of the employees. The study then employed the simple random sampling technique to collect data from 124 hospital staff using questionnaires. The main data collection instrument for the current study was questionnaires. A pilot-test was conducted in Kapsabet Referral Hospital to determine the validity and reliability of the research instruments. To establish the validity of the data collection instruments, the research instruments were given to supervisors and experts in the field of human resources. The reliability of the instrument was tested through the use of the Cronbach Alpha value, that is, to establish the reliability of the questionnaire. Data analysis was done using descriptive and inferential statistics. Descriptively, data was analyzed using frequency, percentages, means, and standard deviations. Inferentially, data was analyzed using correlation and multiple regression models. The analyzed data was presented in the


Introduction
Human resource (HR) had many important functions in the organization.These included recruitment, performance management, learning and development, and many more (Whysall, Owtram & Brittain, 2019).In total, there were 12 key functions of Human Resources Management.HRM was a distinctive approach to employment management that sought to achieve competitive advantage through the strategic deployment of a highly committed and capable workforce using an integrated array of cultural, structural, and personnel techniques (Bradley, 2016).Performance management was also an instrument to close the gap between the workforce they had at that time and the one they wanted to have in the future by helping employees develop future-ready skills and competencies (Ahammad, 2017).Good documentation practices were guidelines for document management and control designed to help companies in regulated industries meet required quality standards.Good documentation were guidelines for document management and control designed to help companies in regulated industries meet required quality standards (Penn & Pennix, 2017).They made sure the department was compliant with both so the company did not run into legal troubles.That's why eFileCabinet allowed them to automatically set who could see or manipulate documents that they had on file.That way, they only had to set those permissions once, and eFileCabinet would set those permissions automatically for them every time (Motamed-Khorasani, 2017).Human resource management in Malaysia became a highly discussed topic when the significance of talent management was shown regardless of company size nor type of industry (Yusliza et al., 2020).A number of studies suggested that effective human resource management could be the main factor for the success and sustainability of a company.Nevertheless, human resource management in Malaysia yielded companies' attention due to the fact that job-hopping frequently occurred in today's hiring market (Chams & García-Blandón, 2019).Malaysian companies actively sought solutions to make good employees stay in the company by enhancing employee's motivation, furthermore, to reduce the turnover rate by improving human resource management (Abdullah & Abubakar, 2017).In Nigeria, the discipline of Human Resource Management witnessed a great deal of change over the past decades, and these changes represented two major transformations (Castles, 2016).The first was the transformation from being the field of personnel management to being the field of human resource management.The second was the transformation from the field of HRM to the field of strategic human resource management (Fahim, 2018).The second transformation was based upon the recognition that, in addition to aligning human resource policies and practices with each other, they needed to be linked with the needs of the organization.Given that these needs were reflected in the strategies of the firm, this transformation of "human resource management" came to be known as "strategic human resource management" (Bibri, 2021).
In Kenya and other developing countries with similarly complicated HR scenarios, two fundamental problems affected the determination and direct application of HRM improvements to the health system (Almaaitah, Harada & Almaaitah, 2017).In such countries, the health system was not a single, cohesive organization with its own integrity, at least as far as HR policies and practices were concerned.HRM was a system that worked best when the "target" was known and it was an autonomous entity.When implementers were faced with a fragmented public health system that had no authority over basic HR functions and fairly limited HR "decision space," it was often unclear where and how HRM should be anchored as a strategic organizational activity (Muthathi, Levin & Rispel, 2020).

Theoretical Framework The Life Cycle Model
The life cycle model for managing records, as articulated by Theodore Schellenberg and others, has been the prominent model for North American archivists and records managers since at least the 1960s.This model portrayed the life of a record as going through various stages or periods, much like a living organism.In stage one, the record was created, presumably for a legitimate reason and according to certain standards.In the second stage, the record went through an active period when it had maximum primary value and was used or referred to frequently by the creating office and others involved in decision making.During this time, the record was stored on-site in the active or current files of the creating office.At the end of stage two, the record could be reviewed and determined to have no further value, at which point it was destroyed, or the record could enter stage three, where it was relegated to a semi-active status, which meant it still had value but was not needed for dayto-day decision making.Because the record did not need to be consulted regularly, it was often stored in an off-site storage center.At the end of stage three, another review occurred, at which point a determination was made to destroy or send the record to stage four, which was reserved for inactive records with long-term, indefinite, archival value.This small percentage of records (normally estimated at approximately five per cent of the total documentation) was sent to an archival repository, where specific activities were undertaken to preserve and describe the records.The Life Cycle Model made several assumptions.For example, the theory assumed that people depleted their wealth during old age.Often, however, the wealth was passed on to children, or older people may have been unwilling to spend their wealth.The theory also assumed that people planned ahead when it came to building wealth, but many procrastinated or lacked the discipline to save.Another assumption was that people earned the most when they were of working age.However, some people chose to work less when they were relatively young and continued working part-time when they reached retirement age.As a result, one implication was that younger people were more able to take on investment risks than older individuals, which remained a widely accepted tenet of personal finance.The life cycle model not only described what would happen to a record, but it also defined who would manage the record during each stage.During the creation and active periods, the record creators had primary responsibility for managing the record, although records managers may have been involved to various degrees.In the semi-active stage, it was the records manager who took center stage and assumed major responsibility for managing the records.Finally, in the inactive stage, the archivist took the lead in preserving, describing, and providing access to the archival record.The theory showed that individuals planned their spending over their lifetimes, taking into account their future income.Accordingly, they took on debt when they were young, assuming future income would enable them to pay it off.They then saved during middle age in order to maintain their level of consumption when they retired.

Methodology
The study adopted a descriptive survey design.Descriptive survey design was all about describing people who took part in the study.In survey method research, participants answered questions administered through interviews or questionnaires.After participants answered the questions, researchers described the responses given.Descriptive survey design was adopted to determine whether using record-keeping practices was an effective method for the performance of human resource functions through asking questions to the carefully selected sample.This was because it made use of both qualitative and quantitative data to describe the state of affairs as they existed in the field.This design was simple and easy to carry out, yet it could yield suitable information desirable by the study (Mugenda & Mugenda, 2003).Descriptive studies were more than mere data collection; they involved measurement, classification, analysis, comparison, and interpretation of data (Kothari, 2009).Descriptive survey design was useful in the collection of original data from a population that was too large to observe directly.The descriptive analysis approach was chosen for the study because it sought to gain some insight into a phenomenon as a means of providing basic information in the area of study.The target population referred to the individuals or elements that the researcher intended to use in the study in order to obtain primary data (Etikan Musa & Alkassim, 2016).The target population entailed employees of MTRH.The accessible population for the study was; 1 human resources manager, 3 benefits administrators, 18 health administrative officers, 86 training and development manager, 6 recruiter, 1 employee relations specialist and 65 HRIS Specialist.Therefore, the total accessible population was 180 respondents.2012), a sample size was part of the target population that had been procedurally selected to represent the entire population.The sample size of this study was drawn using the Yamane formula for determining the sample size, which was given by: Where n= corrected sample size, N = population size, and e = Margin of error (MoE), e = 0.05 based on the research condition.n=180/ (1+180*0.05 2 ) n=124 Thus, the sample size of the study was 124 respondents.The sample size was obtained using the stratified sampling technique with the strata representing the cadre of the employees.The sample for each stratum was obtained using proportionate stratification.The study then employed the simple random sampling technique to collect data from 124 hospital staff using questionnaires.According to Etikan, Musa, and Alkassim (2016), simple random sampling was appropriate when all the sample population had an equal chance of being chosen to participate in the study.This technique was appropriate for this study because it allowed avoiding bias in selecting respondents and was easy to conduct.The main data collection instrument for the current study was questionnaires.In carrying out this study, the researcher developed a questionnaire of closed-ended type to be used to collect data from the respondents.The closed-ended questionnaires were a Likert scale (1-5), where 5 represented strongly agree, 4 represented agree, 3 represented undecided, 2 represented disagree, and 1 represented strongly disagree.The questionnaires comprised investigative questions and personal details.It was the most convenient instrument for such a big population and enabled respondents to express their views freely concerning records keeping practices and the performance of human resource functions (Alshenqeeti, 2014).The questionnaires were used in this study since they saved time and could enable information to be gathered within the shortest possible time.They facilitated the collection of potential information from a large sample of respondents, and they enabled the responses to be gathered in a standardized way.A pilot-test was conducted in Kapsabet Referral Hospital, where 12 questionnaires were issued out representing 10% of the sample size (Zohrabi, 2013).The researcher carried out a pilot test before the final and actual data collection process.Hofisi, Hofisi, and Mago (2014) prescribed that a pre-test sample should be between 10% and 30%, depending on the sample size.The pilot study assisted in determining the validity and reliability of the research instruments and the operational aspects of administering the questionnaires.In this study, the purpose of a pilot test was to discover possible weaknesses, inadequacies, ambiguities, and problems in any aspect of the research process.Validity was often defined as the extent to which the instrument measured what it purported to measure (Cohen, Manion, & Morrison, 2013).For validity to take place, an instrument should have been reliable for better results, though an instrument could have been reliable as required but not necessarily valid.According to (Frels & Onwuegbuzie, 2013), validity was the degree by which the sample of test items represented the content the test was designed to measure.To establish the validity of the data collection instruments, the research instruments were given to supervisors and experts in the field of human resources.The researcher used content validity to analyze whether the instruments answered the research questions since it was a measure of the degree to which data collected using a specific instrument represented a specific domain or content of a specific concept.Reliability was the ratio of the true score's variance to the observed variable's variance.It could estimate and evaluate the stability of measures and internal consistency of the measurement instrument while rating the reliability of the instrument scores (Lewis, 2015).Reliability meant the ability of a measuring instrument to give accurate and consistent results.According to Creswell (2013), it was necessary to have a pilot test for testing the reliability of data collection instruments.Reliability was the consistency of measurement, which was assessed using the test-retest reliability method.This study tested internal consistency reliability.Sijtsma (2009) defined internal consistency as the extent to which all of the items of a test measured the same construct, that is, the general factor saturation.The reliability of the instrument was tested through the use of Cronbach Alpha value.That is, to establish the reliability of the questionnaire.Cronbach alpha coefficients were reported as an indication of the construct reliability of the measuring instruments.Values ranged from 0 to 1, with higher values indicating greater reliability.An alpha coefficient of below 0.60 was unacceptable, between 0.60 and 0.65 undesirable, between 0.65 and 0.70 minimally acceptable, between 0.70 and 0.80 respectable, between 0.80 and 0.90 very good, and > 0.90 was considered perfect.However, if a Cronbach Coefficient alpha of α = 0.70 was obtained, then it indicated that the research instruments were reliable and therefore could be adopted for data collection.The researcher first obtained an introduction letter from Jomo Kenyatta University of Agriculture and Technology, which allowed the researcher to proceed for data collection as well as provided details of the researcher and the purpose of the study.Upon receiving the introduction letter, the researcher then proceeded to the institution to seek permission to carry out the proposed research.The researcher explained to the respective respondents the purpose of the study and sought cooperation from them.The researcher administered the questionnaire copies personally.The collected data was cleaned, coded, managed, and analyzed with the aid of SPSS software version 23.Data analysis was done using descriptive and inferential statistics.Descriptively, data was analyzed using frequency, percentages, means, and standard deviations.Inferentially, data was analyzed using correlation and multiple regression models.The regression model used to test the hypotheses is shown below:  =  +     + ԑ  …………………………Equation 1  represent performance of the Human resource functions α represent constant.β1, represent the slope which represents the degree in which performance of the Human resource functions changes as the independent variable change by one-unit variables.  represent HR record documentation practices Analyzed data was presented in form of frequency tables.The following were the assumptions that the data had to meet in order to conduct a linear regression analysis.Normality: It was assumed that the residuals of variables were normally distributed.That is, the errors in the prediction of value Y (the dependent variable) were distributed in a way that approached the normal curve.The normality of distribution was also checked using the Kolmogorov-Smirnov test.Linearity: Linear regression required the relationship between the independent and dependent variables to be linear.It was also important to check for outliers since linear regression was sensitive to outlier effects.The linearity assumption could be best tested with scatter plots.Multicollinearity: There should have been no perfect linear relationship between two or more of the predictors.So, the predictor variables should not have correlated too highly.This was tested using the variance inflation factor (VIF) and tolerance.The variance inflation factor (VIF) measured the impact of multicollinearity among the variables in a regression model.Green (2000) concluded that even though there was no formal criterion for determining the bottom line of the tolerance value or VIF, tolerance values that were less than 0.1 and VIF greater than 10 roughly indicated significant multicollinearity.This conclusion was supported by Tavakol andDennick (2011) andGujarat (2009).This study carried out a multicollinearity test among the variables of the study.Serial correlation/autocorrelation: Gujarat ( 2009) and Cameron ( 2005), both cited in Keraro ( 2014), defined autocorrelation as the correlation between members of a series of observations ordered in time or space.A Durbin-Watson test was used to detect the presence of autocorrelation between the variables of this study.According to Gujarat ( 2009), the Durbin-Watson statistic ranged in value between 0 and 4. A value near 2 indicated nonautocorrelation, a value closer to 0 indicated positive correlation, while a value closer to 4 indicated negative correlation.An autocorrelation test was carried out among the variables of the study.

Findings
Response Rate Questionnaires were distributed to 124 to the respondents.A total of 113 questionnaires were reasonably and adequately completed representing 91.1% response rate (Table 4.1).This response rate was deemed satisfactory as suggested by Field (2013) who recommends 75% as a rule of the thumb for minimum responses.Further, regarding the works of Jaworski and Kohli, 1993;Patra et al. (2010), this response rate is considered satisfactory.Descriptive statistics on HR record documentation practice.Employees were asked to score their level of agreement with the statement on HR record documentation practice and the average response rate and frequency of agreement were calculated.Results are shown in Table 5 Key: For the sake of this chart, SD means Strongly Disagreed, D means Disagree, N means Neutral, A means Agree, and SA means Strongly Agree.4.7 findings indicates that 67(59.3%) of the respondents agreed and 31(27.4%) of the respondents disagreed that organizational information and records are well protected.More, the study's findings revealed that in terms of mean and standard deviations that the organizational information and records are well protected (mean=3.40 standard deviation=1.533).Touray (2021) reviewed that the success of any organization depends on effective records management practice that ensures the right records are available at the right time for effective business operations.The need for proper records keeping is indisputable it is an ordinary and necessary component of virtually all business operations.Furthermore 71(62.8%) agreed and 30(26.6%)disagreed that the organizational documentation meets the compliance standards of operation.In terms of mean and standard deviations that the organizational documentation meets it is compliance standards of operation (mean=3.52,standard deviation=1.363).Barker and Barker, (2018) agrees that documents aims to assist organizations in establishing effective key management practices to ensure the confidentiality, integrity, and availability of their cryptographic keys.It provided with recommendations and guidelines for key management processes, roles and responsibilities, policy development, and key life cycle management.Further, 74(65.5%) of the respondents agreed and those who disagreed 20(17.6%)that Information documented are easily accessed by relevant authorities.Furthermore, the study's findings revealed that participants agreed (mean=3.71,standard deviation=1.251)that Information documented are easily accessed by relevant authorities.Finally, study further revealed that 66(58.4%) of the respondents agreed and 26(23%) of the respondents disagreed that the information is well retained.In terms of mean and standard deviation that the information is well retained is (mean=3.53,standard deviation =1.247.However, the findings by Sjöberg, Edberg, Rasmussen and Beck, (2021) examined the documentation of end-of-life care for older individuals receiving specialized palliative care, conducted a retrospective review of patient records to gain insights into the documentation practices and the content of the documentation related to end-of-life care.

Inferential Analysis Results
Inferential analyses used in this section were correlation and multiple regression models.Correlation and multiple regression analysis showed the relationship between independent variables and the dependent variable.

Correlation Analysis Results
Pearson correlation analysis was carried out to show the strength and direction of the association between independent and dependent variables.Table 9 presents the results.The study findings in Table 4.17 indicated that HR record documentation practice and performance of the Human resource function had a positive strong and statistically significant correlation (r= 0.783; p<0.01).

Simple Linear Regression Model of HR Record Documentation Practice
The simple linear regression analysis models the relationship between the dependent variable performance on human resource function and independent variable HR record documentation practice.The results are shown in the section that follows; .76952The coefficient of determination (R 2 ) and correlation coefficient (R) shows the degree of association between HR record documentation practice and performance on human resource function.The results of the linear regression in Table 4.18 indicated that R 2 =0.613 and R = 0.783.R value gives an indication that there is a strong linear relationship between HR record documentation practice and performance on human resource function.The R 2 indicates that explanatory power of the independent variables is 0.613.This means that about 61.3% of the variation of performance on human resource function is explained by the regression model.This implies that the HR record documentation practice had an influence on performance of human resource function at Moi Teaching and Referral Hospital Eldoret.These results are consistent with the study by Chirchir, Aruasa and Chebon, (2021) who emphasize showed that there is a statistical and significant linear connection between record documentation practice and HR function.10 the F test provides an overall test of significance of the fitted regression model.The F value indicates that all the variables in the equation are important hence the overall regression is significant.The F-statistics produced (F = 175.707)was significant at p=0.000 thus confirming the fitness of the model and therefore, there is statistically significant relationship between HR record documentation practice on performance of human resource function.The study results in Table 11 revealed that there was positive linear effect of HR record documentation practice on performance of the Human resource function (β1=.691,p=0.000).This reveals that an increase in HR record documentation practice increases performance of the Human resource function by 0.813 units.Y =1.490 +0.691X1 ………………………………………………….….Equation 2Hypotheses Testing From the regression model computed in Table 11, the research hypotheses were tested using the significance level of the coefficients.The research aimed to test the hypothesis with an aim of failing to reject or rejecting the relationship between independent and the dependent variables.The research hypothesis for the study included; H01: HR record documentation practice has no significant influence on performance of the Human resource function in MTRH.The regression results in Table 4.32 indicate that there is significant relationship between HR record documentation practice and performance of the Human resource function with a beta coefficient of .691and significance of (p= 0.000).The study rejected the hypothesis.

Table 15 Summary of Hypotheses Test Results
Hypotheses coeff P-value Decision H01; HR record documentation practice has no significant influence on performance of the Human resource function in MTRH. .

691
.000 Rejected the null hypothesis

Conclusion of the Study
The study concluded that information documented are easily accessed by relevant authorities on the HR record documentation practice on the performance of the Human resource function at Moi Teaching and referral hospital This implies that planning documentation is well established in the process of project management.Good strategy and policy formulation systems in the management

Recommendation of the Study
The study recommended that the management Moi Teaching and referral hospital should always have a proper planning of documentation which is well established.This will ensure that project evaluation is well achieved.