Sage Journals: Discover world-class research

Abstract

In this study, research materials for the spatial concentration and utilization sharing of research equipment were surveyed, and then the effect factor and the effectiveness were analyzed based on the research results. Also, information regarding research result creation was provided through research equipment utilization sharing. The research results of 100 researchers (25 national research institutes in the Ministry of Science, ICT (Information and Communications Technologies), and Future Planning of South Korea) were chosen for the effect analysis. For the study results, the medicine and pharmacy researchers showed better performance in research equipment utilization sharing than the natural science and engineering researchers. The number of research paper coauthors and the research equipment utilization sharing execution policy of research institutes influenced the rate of research equipment utilization sharing. The research field, the number of research paper coauthors, the research equipment utilization sharing execution policy of research institutes, and the research institute characteristics influenced the utilization sharing of research equipment in the research environment. Also, the utilization sharing of research equipment was statistically significantly influenced by the number of research papers and the impact factor (IF). The utilization sharing of research equipment was not statistically significantly influenced, however, by the IF mean. In this study, the quantitative performance index was found to be effective, and the qualitative performance index was found to be ineffective. In the analysis model, when the researchers followed the research equipment utilization sharing execution policy of the research institutes, the research results improved.

Keywords

Research result research equipment utilization sharing spatial concentration national research institute South Korea science and technology analysis science research productivity research paper research fund research project

Introduction

The national research facility and equipment budget in the South Korean government’s research and development (R & D) budget (US$5,855,883,611) for 5 years (2010–2014) was 5.6% (US$327,929,482). The Ministry of Science, ICT (Information and Communications Technologies), and Future Planning of South Korea continually tried to expand the research facility and equipment investment (2013, 4.8% → 2014, 3.9%) and the research equipment utilization sharing acceptance (2013, 59.7% → 2014, 60.5%). The said ministry has actively embarked on effective investment in and utilization sharing of the national research facilities and equipment (National Science & Technology Council, 2015). Also, the national research facilities and equipment have been opened to public use and sharing to produce synergy and for growth promotion. Research equipment are classified into solo research equipment and research equipment for utilization sharing. The research equipment for utilization sharing are classified into the general research equipment, sectorial research equipment, and exclusive research equipment. The research equipment spatial concentration model is shown in Figure 1. The spatial concentration classification system of research equipment is presented in Table 1. The national research institutes of South Korea have spatially concentrated the country’s research equipment through the corresponding annual plan. The Ministry of Science, ICT (Information and Communications Technologies), and Future Planning of South Korea selected excellent institutes for the utilization sharing service of research equipment, and then the selected institutes were designated as national research equipment centers in 2015 and have been serving as such ever since. Also, Open Lab construction has been promoted through the consigned management expansion of the enterprise and the service-oriented design. The research equipment utilization capacity service is deliberated on and is reflected in the research plan in the R & D planning. Integrated procurement of research equipment has been performed based on the results of a research equipment demand survey. Also, the exclusive organization of research equipment and manager appointment is needed in research institutes. Spatial concentration opening and research equipment utilization sharing are needed to produce synergy and for R & D growth promotion. For these, research institutes need research operation that is committed to the research institute of the service enterprise of the research equipment. Also, research institutes need high-quality test and data analysis service provision.

Figure 1.

Research equipment spatial concentration model.

Table 1.

Spatial concentration classification of research equipment.

	Research equipment utilization sharing
Use type	General research equipment (A)	Sectorial research equipment (B)	Exclusive research equipment (C)
Installation unit	Central center	Sectorial laboratory	Laboratory (utilization sharing)
Use scope	Whole institute + outside	Institute sector + outside	Institute sector + outside

In this article, the effectiveness of the research results for the research institute and the researcher was verified through the analysis of the effect of the research results on research equipment spatial concentration and utilization sharing. Also, the contribution of the research results and the research productivity to the utilization sharing of research equipment was confirmed. This article proposes research support policy implications for the utilization sharing of research equipment. This research has academic value owing to its analysis of the effect of research results and its data survey on the spatial concentration and utilization sharing of research equipment. Information was provided for research result creation through the utilization sharing of research equipment.

Conceptual research and analysis framework for research result management

The purpose of the analysis of research results is to conduct a survey on the quantitative performance, qualitative performance, and influence of the research results.¹ It includes the following questions: How many research results are going to be created? How many research results are more excellent compared to those of other research institutes? Also, research results could provide objective analysis data to policy makers, decision makers, and evaluators for research support and fairness of evaluation. The important purpose of the provision of research results is to provide technology trend information to researchers for research objectivity and direction setting.² Research result analysis is classified into elemental analysis and in-depth analysis. Elemental analysis provides technology trend information and the statuses of paper publications for major countries and research institutes, of collaborative research for major countries, of coauthors, and of major journals, among others, through descriptive statistics and citation analyses.³ The research productivity (of a country, an institute, and a researcher) can be analyzed through the number of paper publications.⁴ In-depth analysis analyzes the qualitative performance with regard to the influence and the ripple effect through citation analysis, citation mining, and social network analysis. The journal impact factor (JIF) and the citation index are utilized for the influence level analysis of research results.^5,6 The analysis framework of research results is shown in Figure 2.

Figure 2.

Research result analysis framework.

Research result management

Research result management is focused on intellectual property (the research report, research paper, patent, etc.).⁷ It discusses the securing, maintenance control, and utilization of intellectual property and is applied to the market efficiency principle for the government sector as well as to result-oriented management rather than investment management in the R & D policy. In general, the concept of research result including the valuable and creative knowledge is created on the research process. The valuable and creative knowledge can be openly utilized. The output of research result consists of the patent, the paper, and the product. Recently, the research result includes the outcome for the economic effect, the social effect, the scientific and technological innovation, and so on. The research result can create the new value through the R & D diffusion, the implementation, and the commercialization. The research result management can expand the R & D diffusion through the transaction, the transfer, and the distribution for research result. The market efficiency principle for the government sector as well as of result-oriented management is becoming more and more important because the research result management is aimed at the economic and social utilization of research result.^8

–11

Research result evaluation

Research result evaluation is focused on research project evaluation.¹² It considers the research objectives, expected results, and research result evaluation method of research projects. Also, it has been used for the measurement and evaluation of the mission accomplishment of research institutes and for the attainment of the objectives of R & D projects.¹³ It is currently being actively applied to R & D program evaluation. In particular, economic-result measurement research has been performed in the OECD (Organization for Economic Cooperation and Development) countries for objective result measurement.¹⁴

Research productivity index

In general, the research productivity index has been used for the quantitative analysis of research inputs and outputs.¹⁵ Research results (papers, conference presentations, etc.) are surveyed, after which the ratio of data types is confirmed. Also, paper publication results are analyzed in terms of the research fund (the research cost support organization, the government research cost, the R & D cost, etc.). Most researchers announce the research fund, which is acknowledged by the research paper. The representative index is being used for the paper publication number analysis for the science citation index (SCI) journal of Thomson Reuters.¹⁶ The number of paper publications of SCI can be analyzed through the research institute announcements.¹⁷ The journal-to-field impact score (JFIS) can be used for the analysis of the qualitative productivity of the journal and of the number of papers published by the journal. Below is the detailed formula.

JFIS= \frac{JI}{FI}

where FI is the mean citation number for publication journal J of Y year and JI is the mean citation number for identical field journals of publication journal J of Y year.

The JIF is used for measuring research results through journal paper number analysis (the top rank or the ranks in the top 5%). Below is the detailed formula.

P_{f 5 %} = \frac{(\frac{P_{f 5 %}}{P})}{u_{f 5 %}}

where P_f _5% is the citation paper number of the rank in the top 5%, u_f _5% is the citation paper number of the rank in the top 5%/total paper number, and P is the total paper number.

The activity index (AI) is used for the comparative analysis of the research institute level and the global average R & D and technology level. Below is the detailed formula.

AI = \frac{(\frac{P_{n}}{A_{n}})}{(\frac{P}{A})}

where P_n is the paper number of research institute, A_n is the author number of research institute, and P is the total paper number.

The relative AI (RAI) can be obtained through a detailed analysis of the research field level and the world standard. Below is the detailed formula.

RAI = \frac{(\frac{P_{r d}}{P_{r}})}{(\frac{P_{w d}}{P_{w}})}

where P_rd is the paper number of research institute for detail research field, P_r is the paper number of research institute for research field, P_wd is the paper number of world for detail research field, and P_w is the paper number of world for research field.

Citation index

The citation index is widely utilized for quantitative and qualitative analyses. In general, the citation index is high for old research papers and differs from one research field to another.^18,19 Therefore, it is utilized for basic data research. Relative comparison is better than using the absolute standard for analysis of research results. The citation index is used for the normal analysis of research results.^20,21

The citations per publication (CPP) refers to the mean influence of the research paper. Below is the detailed formula.

CPP = \frac{C}{P}

where C is the total citation number and P is the total paper number.

The field citation score (FCS) is utilized for the research institute level and the world average. Below is the detailed formula.

FCS = \frac{C_{s}}{P_{s}}

where C_s is the total citation number of specific field and P_s is the total paper number of specific field.

The influence can be measured through the mean citation index of research papers. The field citation mean (FCSm) can be calculated through the citation number of the field concerned. Below is the detailed formula.

FCSm = \frac{(\frac{C_{s}}{P_{s}})}{(\frac{C}{P})}

where C_s is the total citation number of specific field, P_s is the total paper number of specific field, C is the total citation number, and P is the total paper number.

The h-index can be considered reflective of both the quantity and quality of a research institute’s research papers. It is the point at which the citation number is higher than or the same as the research paper number. The conceptual diagram of h-index is shown in Figure 3. The JIF indicates the relative importance of a journal for the research field. Y refers to the year. Below is the detailed formula.

Figure 3.

Conceptual diagram of h-index.

JIF = \frac{CI}{PI}

where CI is the total citation number of Y year paper for Y – 2 year and Y – 1 year and PI is the total paper number for Y – 2 year and Y – 1 year.

The journal citation score (JCS) is utilized to establish the result level of a research paper and is determined through the influence analysis of the research paper. Below is the detailed formula.

JCS = \frac{(\frac{C_{r}}{P_{r}})}{(\frac{C_{a}}{P_{a}})}

where C_r is the total citation paper number of A journal for research institute, P_r is the total paper number of A journal for research institute, C_a is the total citation paper number for A journal, and P_a is the total paper number for A journal.

The rank-normalized IF (RNIF) can be determined through the journal rank and influence analysis.²² Below is the detailed formula.

RNIF = \frac{(N_{j} - R_{i} + 1)}{N_{j}}

where N_j is the journal number of research field and R_j is the journal JIF rank of research field.

The scope-adjusted IF (SAIF) was devised for the fair evaluation of specific field journals. When the citation index of a specific field journal is high, the SAIF declines. Below is the detailed formula.

SAIF = \frac{JII}{CPN} \times 1000

where JII is the journal influence index and CPN is the citation paper number.

The citation index of a specific field journal and the citation index of another same-field journal are compared.²¹ Below is the detailed formula.

MCI = \frac{CPP}{FCSm}

where MCI is the mean citation index, CPP is the citations per publication, and FCSm is the field citation mean.

Research result influence and spread effect analysis

The research result influence analysis method utilizes the citation information.^23,24 The paper publications of a journal are surveyed through influence measurement. Also, the qualitative influence is analyzed through excellent paper selection. The research result ripple effect analysis targets the economy, society, culture, technology, and so on and is focused on technologies and studies.²⁵ The IF information is collected for the calculation of the journal’s influence.

Research method for research result effect analysis

Research object and data collection

In this article, the research results of 100 researchers (25 national research institutes in the Ministry of Science, ICT (Information and Communications Technologies), and Future Planning of South Korea) were chosen for effect analysis (2015). All the researchers had a PhD in natural science, engineering, or agriculture, fisheries, and oceanography. This is based on the research result report (2016) of National Research Council of Science & Technology. Data on the research results on the spatial concentration and utilization sharing of research equipment were collected from 25 national research institutes.

Dependent variables

The dependent variables included the number of research papers, the average IF, and the sum of IF, which are presented in Table 2. Such dependent variables can show both the quantity and quality of research results. The number of research papers is a widely utilized variable. In this study, the number of SCI journals was utilized for 3 years (2013–2015). For the research variables, the average IF and the sum of IF was added to the JIF. The IF is the citation index of Thomson Reuters. It was calculated based on the citation number of a research paper for 2 years. The sum of IF indicates both the quantity and quality of research results.

Table 2.

Dependent variables.

Type	Variable	Variable description
Quantitative index	Research paper number	SCI paper total IF (2013–2015)
Qualitative index	Mean IF	Research paper mean IF (2013–2015)
Integration index	Total IF	Research paper total IF (2013–2015)

IF: impact factor; SCI: science citation index.

Independent variables

The independent variables in this study are presented in Table 3. The spatial concentration and utilization sharing of research equipment are defined as the free provision of analysis support and utilization sharing service to researchers. In this study, the spatial concentration of research equipment refers to the number of research paper publications through the utilization sharing of research equipment. The research fields were natural science, engineering, and agriculture, fisheries, and oceanography. The research field was coding in preparation for medicine and pharmacy. The research environment consisted of the research fund and the number of coauthors. The research institute characteristics consisted of the research paper number per researcher, the research project number per researcher, and the research equipment utilization sharing execution policy.

Table 3.

Independent variables.

Classification	Variable	Variable description
Research equipment utilization sharing	Research equipment utilization sharing execution	Research paper publication through research equipment utilization sharing
Research field	Natural science Engineering Agriculture, Fisheries, and oceanography	“1” is a code in preparation for medicine and pharmacy
Research environment	Research fund Coauthor	Total research fund (2012–2014) Mean coauthor number (2013–2015)
Research institute characteristics	Research paper number per researcher Research project number per researcher Research equipment utilization sharing execution policy	Mean research paper number (2013–2015) Mean research project number (2013–2015) 0 = Nonexecution policy 1 = Execution policy

Two-stage least squares estimation using an instrumental variable

In this article, a verification model is proposed for the effect analysis of the research results on the spatial concentration and utilization sharing of research equipment. Below is the detailed formula for the spatial concentration and utilization sharing of research equipment.

U = α_{0} + α_{i} \sum_{1}^{3} R_{f} + α_{i} \sum_{4}^{5} R_{e} + α_{i} \sum_{6}^{7} R_{c} + α_{8} (U_{p}) + u

where U is the utilization sharing of research equipment, R_f is the research field, R_e is the research environment _i , R_c is the research institute characteristics _i , and U_p is the utilization sharing execution policy of research equipment.

Below is the detailed formula for the ordinary least squares (OLS).

R = β_{0} + β_{i} \sum_{1}^{3} R_{f} + β_{i} \sum_{4}^{5} R_{e} + β_{i} \sum_{6}^{7} R_{c} + β_{8} (U_{e}) + e

where R is the research result, R_f is the research field _i , R_e is the research environment _i , R_c is the research institute characteristic _i , and U_e is the utilization sharing execution of research equipment.

When the regression coefficient of the variable is unreliable, two-stage least squares (2SLS) estimation is utilized through the instrumental variable. In 2SLS, the instrumental variable acts independently through one-stage least squares estimation. Also, the research equipment utilization sharing execution is limited to the external variation. It can explain the correlation of the error term (e) for the instrumental variable. Below is the detailed formula.

R = ∊_{0} + ∊_{i} \sum_{1}^{3} R_{f} + ∊_{i} \sum_{4}^{5} R_{e} + ∊_{i} \sum_{6}^{7} R_{c} + ∊_{8} (U_{e}) + v

The instrumental variable needs the correlation for the endogenous explanatory variable. When the F verification statistic is larger than 10, it is the proper correlation in the performance results of SLS 1 phase.²⁶ The F verification statistic of Wu–Hausman is proposed for the endogenous explanatory variable.²⁷

LSDV fixed-effect model

The least squares dummy variable (LSDV) is a statistical model that represents the observed quantities in terms of explanatory variables that are treated as if the quantities were nonrandom. This is in contrast to random effects models and mixed models in which either all or some of the explanatory variables are treated as if they arise from random causes. This model is called the fixed-effect model.^28

–31,32 The panel data analysis is a statistical method, widely used in social science, epidemiology, and econometrics, which deals with two and n-dimensional panel data. In the model of panel fixed effect, the term fixed effects estimator is used to refer to an estimator for the coefficients in the regression model.³³ In this article, the panel data and the mean were utilized for the research paper publication period. The LSDV fixed-effect model was utilized for the effect analysis of the research results for the spatial concentration and utilization sharing of research equipment. Therefore, the LSDV was utilized. Below is the detailed formula.

y_{t i} = \sum_{i = 1}^{100} α_{i} + \sum_{k = 1}^{3} β_{k t i} + e_{t i}

Research result effect analysis

The descriptive statistical analysis results for the research field are presented in Table 4. They show that there are 39 researchers (39%) in the engineering field, 35 (35%) in the medicine and pharmacy field, 19 (19%) in the natural science field, and 7 (7%) in the agriculture, fisheries, and oceanography field. The descriptive statistical analysis results for the research results, research environment, and research institutes, on the other hand, are presented in Table 5. They show that the researchers presented 15.3 papers on average for 3 years (2013–2015). This means that the researchers presented 5.1 papers on average for 1 year. The mean IF was thus 2.8, which is the average number of research papers completed by the researchers for 3 years (2013–2015). For the research fund, a log value was utilized for the actual analysis. The total research fund was US$6,051,000 on average for 3 years (2012–2014) and US$410,000 on average per year. The mean number of coauthors per year was 6.1 persons. The average number of research papers per researcher per year was 0.59, and the average number of research projects per researcher per year was 1.61. The cross-analysis results for research equipment utilization sharing and for the research equipment utilization sharing execution policy are presented in Table 6. They show that there were 82 researchers (82%) who published a research paper through research equipment utilization sharing. There were 77 researchers who officially claimed to have engaged in research equipment utilization sharing and 23 who claimed not to have done so. Based on the analysis results for the research equipment utilization sharing execution policy, it was found that the said policy promotes the utilization sharing of research equipment. The results of the mean difference analysis for the utilization sharing of research equipment are presented in Table 7. The research paper number, the mean IF, and the total IF had significant differences in terms of research equipment utilization sharing. The research equipment utilization sharing execution group had higher mean research result index values than the research equipment utilization sharing nonexecution group. Moreover, the research paper number and total IF values obtained by the research equipment utilization sharing execution group were twofold higher than those obtained by the research equipment utilization sharing nonexecution group. The difference in the mean IF value, however, was small. Based on these results, it can be said that the research equipment utilization sharing execution influenced the research results. There were significant differences between the natural science and engineering fields in terms of the independent variable values. This means that the variables have correlations. Also, the selection bias possibility can be suggested for research equipment utilization sharing. 2SLS was utilized for accurate analysis. The linear probability model and the probit for research equipment utilization sharing are shown in Table 8. Among the research fields studied, the correlation between the natural science and engineering fields was statistically significant. Also, such fields showed a strong possibility of being able to execute research equipment utilization sharing. In the agriculture, fisheries, and oceanography field, the rate of research equipment utilization sharing was lower, but the difference was not statistically significant. The number of coauthors showed a statistically significant correlation with the research execution. The research fund, research paper number per researcher, and research project number per researcher showed statistically insignificant correlations with the rate of research equipment utilization sharing. With regard to the research institute characteristics, the research equipment utilization sharing execution policy was a strong variable. The results of this study have academic value in that they verify the influence variable of research equipment utilization sharing.

Table 4.

Descriptive statistical analysis results for the research fields.

Variable	Classification	Number of people	Percentage (%)
Research field	Natural science	19	19
	Engineering	39	39
	Medicine and pharmacy	35	35
	Agriculture, fisheries, and oceanography	7	7
Total		100	100

Table 5.

Descriptive statistical analysis results for the research results, research environment, and research institutes.

Variable	Mean	Standard deviation	Minimum value	Maximum value
Research paper number	15.31	16.53	1	130
Mean IF	2.8	1.71	0.05	9.01
Total IF	38.90	55.10	0.06	431.03
Research fund (US$)	410,000	560,000	0	6,051,000
Coauthor	6.10	2.39	1	16.10
Research paper number per researcher	0.59	0.23	0	1.32
Research project number per researcher	1.61	0.87	0.30	5.10

IF: impact factor.

Table 6.

Cross-analysis results for research equipment utilization sharing and the research equipment utilization sharing execution policy of research institutes.

	Research equipment utilization sharing execution policy
Research equipment utilization sharing	Nothing	Execution	Total
Nothing	6 (26%)	12 (16%)	18 (18%)
Execution	17 (74%)	65 (84%)	82 (82%)
Total	23 (100%)	77 (100%)	100 (100%)

Table 7.

Mean difference analysis results for research equipment utilization sharing.

Classification	Execution (n = 18), Mean (standard deviation)	Execution (n = 82), Mean (standard deviation)	t	Significance level
Research paper number	8.70 (7.01)	13.60 (14.07)	−4.70	0.01
Mean IF	1.91 (1.37)	2.53 (1.64)	−3.01	0.01
Total IF	16.03 (17.14)	39.78 (55.81)	−5.45	0.01
Natural science	0.23 (0.41)	0.20 (0.40)	1.07	0.19
Engineering	0.67 (0.49)	0.36 (0.48)	1.90	0.20
Agriculture, fisheries, and oceanography	0.01 (0.18)	0.04 (0.18)	0.05	0.31
Research fund	3.93 (1.60)	4.85 (2.12)	0.01	0.03
Coauthor mean	3.47 (1.58)	5.82 (2.57)	−3.91	0.01
Research paper number per researcher	3.07 (0.10)	0.43 (0.25)	0.03	0.21
Research project number per researcher	2.59 (0.71)	1.56 (0.84)	0.04	0.56

IF: impact factor.

Table 8.

Linear probability model and probit for research equipment utilization sharing.

Classification	Independent variable	Linear probability model	Probit marginal effect
Research field	Natural science	0.030 (0.010)*	0.100 (0.009)*
	Engineering	0.071 (0.030)**	0.087 (0.030)*
	Agriculture, fisheries, and oceanography	0.053 (0.092)	−0.001 (0.081)
Research environment	Research fund	0.021 (0.010)	0.005 (0.004)
Research environment	Coauthor number	0.001 (0.005)*	0.039 (0.001)*
Research institute characteristics	Research paper number per researcher	0.003 (0.041)	0.010 (0.034)
	Research project number per researcher	−0.023 (0.034)	0.008 (0.009)
	Research equipment utilization sharing execution policy	0.199 (0.001)***	0.138 (0.019)***
Constant		0.709 (0.010)*
F statistic/likelihood ratio χ ²		5.010*	69.100*
Adjusted R ²/pseudo R ²		0.081	0.156

*p < 0.05.

**p < 0.01.

***p < 0.001.

In this study, research equipment utilization sharing was the endogenous explanatory variable, and 2SLS was utilized for endogeneity control. The estimated effects of research equipment utilization sharing on the research paper number are presented in Table 9, along with the OLS analysis results. The F-test result shows the correlation of the endogenous explanatory variable with research equipment utilization sharing. When the verification statistic of the one-stage estimation model was higher than 10, the instrument variable had a correlation with the endogenous variable. In the 2SLS size of the nominal 5% Wald test, the instrument variable exceeds the critical significance level of the weak instrument (10%). The verification result of Wu–Hausman showed endogeneity because it was statistically significant (p > 0.05). As regards the research paper effect of research equipment utilization sharing execution, the 2SLS estimate was statistically significant (p > 0.01). Also, the number of research papers per researcher, the research fund, and the number of coauthors were statistically significantly correlated with the research paper number. The estimated effects of research equipment utilization sharing on the mean IF are presented in Table 10. The OLS value was not statistically significantly correlated with research equipment utilization sharing, but the instrument variable was. As for the regression coefficient, the mean IF of the researchers was very high for the utilization sharing of research equipment. The estimated effects of research equipment utilization sharing on the total IF are presented in Table 11. The F verification statistic of Wu–Hausman was 5.010. It was statistically significant (p > 0.05). The total IF was very high for research equipment utilization sharing. The fixed-effect model was utilized for the robustness panel data analysis of 2SLS estimation. Also, the LSDV fixed-effect model was utilized for the researcher characteristics. The fixed-effect model of research equipment utilization sharing for the research results is described in Table 12. The research paper number and the total IF were statistically significantly correlated with research equipment utilization sharing, the research paper number was increased for research equipment utilization sharing, and the total IF was not statistically significantly correlated with research equipment utilization sharing.

Table 9.

Estimated effect of research equipment utilization sharing on the research paper number.

Independent variable	OLS	2SLS
Research equipment utilization sharing execution	5.010 (1.379)*	18.907 (6.171)*
Natural science	0.010 (0.077)	0.060 (0.019)
Engineering	0.106 (0.001)	0.170 (0.080)
Agriculture, fisheries, and oceanography	0.009 (0.061)	0.070 (0.081)
Research fund	1.503 (0.010)**	1.219 (0.004)***
Coauthor number	1.671 (0.012)**	1.001 (0.001)*
Research paper number per researcher	9.101 (3.780)***	10.101 (3.900)***
Research project number per researcher	0.508 (0.701)*	0.607 (0.511)
Constant	−8.081 (1.117)**	17.080 (4.707)**
F statistic/likelihood ratio χ ²	10.681***	8.810***
Adjusted R ²/pseudo R ²	0.171	0.190
Instrumental variable verification
F test on excluded IV	19.901***
Wu–Hausman F test of exogeneity	5.179

OLS: ordinary least squares; 2SLS: two-stage least squares.

*p < 0.05.

**p < 0.01.

***p < 0.001.

Table 10.

Estimated effect of research equipment utilization sharing on the mean IF.

Independent variable	OLS	2SLS
Research equipment utilization sharing execution	0.313 (0.101)	1.791 (0.509)
Natural science	0.459 (0.139)*	0.380 (0.163)**
Engineering	−0.004(0.130)	−0.016(0.129)
Agriculture, fisheries, and oceanography	−0.009 (0.061)*	−0.510 (0.213)*
Research fund	0.608 (0.010)*	1.017 (0.001)*
Coauthor number	1.190 (0.080)*	1.403 (0.008)***
Research paper number per researcher	0.800 (0.182)***	0.901 (0.351)***
Research project number per researcher	0.031 (0.029)	0.016 (0.031)
Constant	−0.171 (0.014)	−1.069 (0.471)
F statistic/likelihood ratio χ ²	11.040***	18.030***
Adjusted R ²/pseudo R ²	0.407	0.310
Instrumental variable verification
F test on excluded IV	21.001***
Wu–Hausman F test of exogeneity	4.319

IF: impact factor; OLS: ordinary least squares; 2SLS: two-stage least squares.

*p < 0.05.

**p < 0.01.

***p < 0.001.

Table 11.

Estimated effect of research equipment utilization sharing on the total IF.

Independent variable	OLS	2SLS
Research equipment utilization sharing execution	12.301 (6.001)*	51.871 (21.450)*
Natural science	0.459 (0.139)	10.111 (5.031)
Engineering	−1.419 (4.371)	1.617 (5.049)
Agriculture, fisheries, and oceanography	−16.835 (7.190)	−15.081 (6.109)
Research fund	6.500 (1.760)***	3.910 (1.230)***
Coauthor number	8.070 (0.810)***	5.903 (1.411)***
Research paper number per researcher	40.100 (7.001)***	41.090 (9.010)***
Research project number per researcher	3.130 (2.103)	2.910 (1.601)
Constant	−51.004 (8.615)***	−87.001 (16.159)***
F statistic/likelihood ratio χ ²	16.003***	17.701***
Adjusted R ²/pseudo R ²	0.303	0.301
Instrumental variable verification
F test on excluded IV	23.801***
Wu–Hausman F test of exogeneity	5.010*

IF: impact factor; OLS: ordinary least squares; 2SLS: two-stage least squares.

*p < 0.05.

**p < 0.01.

***p < 0.001.

Table 12.

Fixed-effect model of research equipment utilization sharing for research results.

Independent variable	Research paper number	Mean IF	Total IF
Research equipment utilization sharing execution	1.361 (0.181)***	−0.168 (0.149)	4.187 (0.670)***
F statistic/likelihood ratio χ ²	18.54***	30.03***	14.10***
F test, where all u_i = 0	4.056***	2.301***	4.501***
Adjusted R ²	0.401	0.583	0.680

IF: impact factor.

*p < 0.05.

**p < 0.01.

***p < 0.001.

In this article, 2SLS was utilized for the effect and factor analysis of research results for the spatial concentration and utilization sharing of research equipment. The medicine and pharmacy researchers executed research equipment utilization sharing more often than the natural science and engineering researchers did. The number of coauthors of research papers and the research equipment utilization sharing execution policy of research institutes were influenced by the utilization sharing of research equipment. The research field, number of coauthors, and research equipment utilization sharing execution policy of institutes were influenced by the utilization sharing of research equipment in the research environment and by the research institute characteristics. Also, the utilization sharing of research equipment was statistically influenced by the research paper number and the IF but was not statistical significantly influenced by the mean IF. The quantitative performance index was effective, and the qualitative performance index was not. A comparative analysis of OLS and 2SLS estimation was performed for the effect and factor analysis of the research results for the spatial concentration and utilization sharing of research equipment. The analysis model for the instrument variable was more effective than OLS for the utilization sharing of the research equipment. The research equipment utilization sharing execution policy of institutes can greatly improve the research results.

Conclusions and policy implications

In this article, the effect analysis of research results and the characteristics of the research equipment utilization sharing of research institutes are discussed. Conclusions were drawn from the research results and their discussion. The first effect factors of research equipment utilization sharing were the research environment (the research field, research partner, and research equipment utilization sharing execution policy of institutes) and the research institute characteristics. The second effect factors were the research paper number and the IF. Research equipment utilization sharing was not statistically significantly influenced by the average IF. In conclusion, a number of policy implications can be proposed for the spatial concentration and utilization sharing of research equipment. First, a research equipment utilization sharing execution policy can be expected for research result enhancement. Utilization sharing of research equipment in accordance with the government R & D policy is needed for the efficient operation of research equipment. Utilization sharing of research equipment provides more opportunities for the research achievement and the research equipment can be based on the research result. In particular, utilization sharing of research equipment can provide the research opportunity for young researchers, and researchers can study without the research equipment. Also, utilization sharing of research can contribute to an increase in utilization rate and prevents usable idle research equipment and usable underutilized research equipment. Second, an ecosystem and platform can be constructed for connecting the service supply and demand through the utilization sharing of research equipment. This can contribute to generating research results for the R & D innovation system without having to buy research equipment. The R & D budget is reduced and the research efficiency can be increased through the utilization sharing of research equipment. Third, excellent research results can be created through the research equipment utilization sharing of research institutes and through the mutual cooperation among such institutes (test analysis information sharing, conference opening, and researcher networking). This can contribute globally to the research activity promotion, the research result sharing, and the new discovery. This research has academic value through the research result effect analysis and data survey results for the spatial concentration and utilization sharing of research equipment. An attempt at information provision for research result creation was attempted through the utilization sharing of research equipment. In this article, the research field is classified for the natural science, the engineering, the medicine and pharmacy, and the agriculture, fisheries, and oceanography. However, the research field is not evenly distributed. The distribution of research field is the research limitation of this study. In future, research will be conducted on the R & D budget, research network, research innovation cluster, equal distribution of research field, and research equipment platform for the spatial concentration and utilization sharing of research equipment.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research,authorship,and/or publication of this article.

Funding

The author(s) received no financial support for the research,authorship,and/or publication of this article.

References

Greger

. How the citation index identifies highly cited investigators in nutrition. Nutrit Today 2007; 42(5): 223–225.

Dodson

. Citation analysis: maintenance of h-index and use of e-index. Biochem Biophys Res Commun 2009; 387(4): 625–630.

Garfield

. The evolution of the science citation index. Contribut Sci 2009; 5(1): 63–70.

Meho

Rogers

. Citation counting, citation ranking, and h-index of human-computer interaction researchers: a comparison of scopus and web of science. J Associat Inform Sci 2008; 59(11): 1711–1726.

Force

Robinson

. Encouraging data citation and discovery with the data citation index. J Comput Aided Mol Des 2014; 28(10): 1043–1048.

Rodriguez

. The citation indexes and the quantification of knowledge. J Edu Administr 2009; 47(2): 250–266.

Dastidar

. Science citation index, co-citation and the scientists. Curr Sci 2004; 86(5): 626–630.

Buntrock

. Using citation indexes, citation searching, and bibliometrics to improve chemistry scholarship, research, and administration. J Chem Edu 2016; 93(3): 560–566.

. Highly cited Antarctic articles using science citation index expanded: a bibliometric analysis. Scientometrics 2016; 109(1): 337–357.

10.

Hartley

. Classic articles in psychology in the science citation index expanded: a bibliometric analysis. British J Psychol 2016; 107(4): 768–780.

11.

Robinson

Jimnez

Torres

. Analyzing data citation practices using the data citation index. J Associat Inform Sci Technol 2016; 67(12): 2964–2975.

12.

Lederberg

. Early impressions of citation indexing. Curr Sci 2005; 89(9): 101–107.

13.

Wohlin

. A new index for the citation curve of researchers. Scientometrics 2009: 81(2): 521–533.

14.

Lehrl

. Evaluating authors and co-authors with the citation index and impact factor. Dtsch Med Wochenschr 2000; 123(38): 1109–1116.

15.

Force

Auld

. Data citation index: promoting attribution, use and discovery of research data. Inform Serv Use 2014; 34(1): 97–100.

16.

Jacso

. Citation-enhanced indexing/abstracting databases. Online Inform Rev 2004; 28(3): 235–238.

17.

Moed

Hesselink

. The publication output and impact of academic chemistry research in the Netherlands during the 1980’s: bibliometric analyses and policy implications. Res Policy 1996; 25: 819–836.

18.

Ball

Tunger

. Science indicators revisited - Science citation index versus SCOPUS: a bibliometric comparison of both citation databases. Inform Serv Use 2006; 26(4): 293–302.

19.

Kotze

Growing with indexing and the role of citation databases: ISAP shows the way. Indexer 2015; 33(2): 77–83.

20.

Schubert

Braun

. Relative indicators and relational charts for comparative assessment of publication output and citation impact. Scientometrics 1986; 9: 281–291.

21.

Vinkler

. Evaluation of some methods for the relative assessment of scientific publications. Scientometrics 1986; 10: 157–177.

22.

Pudovkin

Garfield

. Algorithmic procedure for finding semantically related journals. J Am Soc Inform Sci Technol 2002; 53(13): 1113–1119.

23.

Kazi

Patwardhan

Joglekar

. Towards a new perspective on context based citation index of research articles. Scientometrics 2016; 107(1): 103–121.

24.

Sivertsen

Larsen

. Comprehensive bibliographic coverage of the social sciences and humanities in a citation index: an empirical analysis of the potential. Scientometrics 2012; 91(2): 567–575.

25.

Leydesdorff

. Mapping interdisciplinarity at the interfaces between the science citation index and the social science citation index. Scientometrics 2007; 71(3): 391–405.

26.

Stock

Wright

Yogo

. A survey of weak instruments and weak identification in generalized method of moments. J Bus Econ Stat 2002; 20: 518–529.

27.

Staiger

Stock

. Instrumental variables regression with weak instruments. Econometrica 1997; 65: 557–586.

28.

Diggle

Heagerty

Liang

. Analysis of Longitudinal Data. Oxford: Oxford University Press, 2002, pp. 169–171.

29.

Fitzmaurice

Laird

Ware

. Applied Longitudinal Analysis. Hoboken: John Wiley & Sons, 2004, pp. 326–328.

30.

Gardiner

Luo

Roman

. Fixed effects, random effects and GEE: What are the differences? Stat Med 2009; 28: 221–239.

31.

Laird

Ware

. Random-effects models for longitudinal data. Biometrics 1982; 38(4): 963–974.

32.

. Comparison of environmental laws publications in science citation index expanded and social science index: a bibliometric analysis. Scientometrics 2016; 109(1): 227–239.

33.

Maddala

. Introduction to econometrics. New York: Wiley, 2001.

The effect analysis of the research results on the spatial concentration and utilization sharing of research equipment

Abstract

Keywords

Introduction

Conceptual research and analysis framework for research result management

Research result management

Research result evaluation

Research productivity index

Citation index

Research result influence and spread effect analysis

Research method for research result effect analysis

Research object and data collection

Dependent variables

Independent variables

Two-stage least squares estimation using an instrumental variable

LSDV fixed-effect model

Research result effect analysis

Conclusions and policy implications

Footnotes

Declaration of Conflicting Interests

Funding

References