Comparison of the results of triangulation of methods for assessing the effectivness of long-term social care institutions in Latvia

Today, countries around the world face a variety of ecological, economic, political and social challenges that are often no longer local within a single country or group of countries today, but are becoming a global problem where smart solutions and planning are needed. One of these challenges is the general ageing of the population, which has also become a topical issue in Latvia, where demand for long-term social care services is increasing. Long-term social care institutions (LTSCI) are experiencing shortages in human resources and financial resources and funding for this is not expected to be further increased. For this reason, it is important to assess the effectiveness of LTSCI economic activity in Latvia in order to optimise the use of available resources and plan for the efficient provision of the service on the basis of the results obtained. This work, using the results of a previous study by authors on the effectiveness of LTSCI resulting from the use of the Data envelopment analysis (DEA) and Stochastic frontier analysis (SFA), reflects the comparison of the results of this triangulation. 156 The aim of the study is to identify whether applying two different research methods to the same selection and performance parameters of LTSCI, the results of which are correlated with each other and thus confirm the objectivity of the assessment of the effectiveness of LTSCI. The previous study carried out by the authors used the 2017 statistics of the Ministry of Welfare of the Republic of Latvia on the registered economic, business, service accessibility and utilised capacity parameters for the activities of 64 Local Municipalities in Latvia which provide long-term social care services for persons of pension age. Of the 36 LTSCI performance parameters recorded in the Ministry of Welfare statistics, the DEA and SFA methods included the 7 most relevant ISAI parameters which, according to the authors, represent the technical and cost-effective performance of LTSCI. Three independent models with different input and output parameters were developed, which were analysed by DEA and SFA methods by establishing 64 LTSCI efficiency ratios (EFRs) for each model. As a result, DEA and SFA methods showed differences between LTSCI EFRs, as the methods include different algorithms for assessing efficiency. The DEA method is based on comparing LTSCI parameters with all other LTSCI parameters, which may result in one or more of the most effective LTSCI attributing (EFR 1). In contrast, SFA efficiency ratios are determined on the basis of the stochastic limits of LTSCI’s operational capabilities, where the effectiveness of the LTSCI is conditional upon the possibility that a particular LTSCI may become more efficient, thus determining the effectiveness of each LTSCI but assigning lower values than (EFR 1). This study therefore compares the EFRs obtained of the three models of the DEA and the SFA three models describing the technical and cost-effectiveness of each individual LTSCI, using Pearson correlation and Spearman rank correlation analysis. Pearson correlation results: Correlation 1 includes coefficient (R) = 0.72 of the DEA and SFA results of the first model of EFRs of DEA and SFA, which describe the technical efficiency of LTSCI and the correlation is positive. Correlation 2 includes DEA and SFA second model EFRs, which represent LTSCI cost-effectiveness and R = 0.58 with positive correlation. Correlation 3 includes DEA and SFA third model EFRs, which represent LTSCI cost-effectiveness and R = 0.72 with positive correlation. Spearman Rank Correlation Results: Correlation 1 includes DEA and SFA first model of EFRs, which describe the technical efficiency of LTSCI, the correlation rate of the results of DEA and SFA methods R = 0.73 with statistical significance (p) < 0.001. Correlation 2 includes DEA and SFA second model EFRs, which represent the cost-effectiveness of LTSCI and R = 0.58 with a positive correlation. As a result of this correlation, R = 0.61 with p <0.001., correlation 3 includes the DEA and SFA third model EFRs, which describe the cost-effectiveness of LTSCI, and R = 0.77 with p < 0.001. Although the EFR rankings of the DEA method differ from the SFA method EFK rankings, the results of both methods show a similar trend, also confirmed by the Pearson and Spearman correlation coefficients and high statistical significance. The work demonstrated the objectivity of LTSCI’s efficiency assessment results on the basis of the application of two different efficiency assessment methods and a strong correlation between the efficiency indicators obtained by these methods.