^{1}

^{2}

^{*}

The public procurement system in Bangladesh has been running by traditional manual process under the flagship of Public Procurement Act (PPA 2006) and Public Procurement Rules (PPR 2008). Public procurement agencies have been facing challenges in this manual tendering system. To overcome this problem and to bring reality to the “Digital Bangladesh” slogan, the Government of Bangladesh introduced the e-Procurement system under the e-GP (Electronic Government Procurement) guideline 2011. After the inception of e-procurement, there is no e-procurement assessment model to improve the e-GP system. The purpose of this research is to develop a conceptual framework and to design an e-procurement assessment model. With this view, we have considered one of the biggest entity Roads and Highways Department (RHD) of the Government of Bangladesh, for field study. Mixed methods along with FGD (Focus Group Discussion), KII (Key Informant Interview), and survey questionnaires are used to collect data from RHD, and then Statistical Package for Social Science (SPSS) software is used for regression analysis and hypothesis testing to develop the e-Procurement assessment model. The novel contribution of the study lies in the test of the hypothesis that focuses on developing the conceptual model of the e-procurement assessment system in Bangladesh. Findings of the study are essential for all Procurement Entity (PE) and suppliers i.e. contractors of RHD who are engaged in the construction of the infrastructure project development project.

Since 2011, Bangladesh has started the e-Procurement system [

According to the seventh five-year plan [

This paper is structured as follows. Section 2 describes related literature, Section 3 provides a conceptual framework for the e-procurement implementation assessment, Section 4 presents the methodology used in this research, and Section 5 contains the key results and discussions. Finally, Section 6 concludes this work.

1) e-Procurement and e-GP Vision

E-procurement [

2) Theoretical Motivation

Every researcher gets important clues designing the research and data analysis technique from the relevant theory. The researcher here studied 2 theories like Maps inititative assessment model and Gardenal e-Procurement assessment model.

a) Gardenal e-Procurement assessment model

Empirical researches on the practical applications, leveraging benefits, and the value of e-Procurement in the public sector of Bangladesh has been done by researchers in a limited way. For the sake of this current review, the model [

b) Maps initiative assessment model

Around the world, Governments are spending about USD 9.5 trillion [

The Methodology for Assessing Procurement Systems (MAPS) is a widely used measurement tool for assessing public procurement in the world which was developed in 2003/2004. Universally accepted MAPS assessments [

MAPS follows [

The following considerations guided the MAPS correction [

• Value for money

• Transparency

• Fairness

• Good governance

MAPS Pillars to assess public procurement are:

i) Regulatory, Legal, and Policy Framework

ii) Institutional Framework and Management Capacity

iii) Public Procurement Operations and Market Practices

iv) Accountability, Integrity, and Transparency of the Public Procurement System.

3) Research Gap Analysis

A critical review of the literature has been made and a summary of the research gap is presented below. In India, the GePNIC system [

The conceptual framework is the output of the in-depth study after a comprehensive critical literature review and research gap analysis on the adoption of e-procurement in an organization of a country. Following is the list of associated variables in

RHD’s bidder and PE community surveys (questionnaires) were considered to measure the significant effects of variables of the conceptual framework. The following 4 hypotheses of

H_{1}.

H_{o}: Effective e-Procurement implementation assessment does not significantly depend on accountability.

H_{a}: Effective e-Procurement implementation assessment significantly depends

Hypothesis No. | Dependent Variables | Independent Variables | Survey Respondents |
---|---|---|---|

H1 | accountability | 1) Collaboration increased between PE and bidders 2) PE officers behavioral change 3) Bidders behavioral change 4) Value for money | PE Officers |

H2 | procurement governance | 1) Capacity development 2) Integrating principles 3) Annual procurement planning and published 4) Quality of govt. tendering process 5) Free from red-tapes 6) Free from problems after contact awarding | PE Officers |

H3 | dematerialization | 1) Paperless procurement process 2) Reduced archiving cost | PE Officers |

H4 | legal, regulatory, and policy framework | 1) Overall CPTU performance 2) PPA 2006 and subsequent amendments 3) PPR 2008 and subsequent amendments 4) e-GP guideline 2011 5) Implementation of digital signature | PE Officers |

on accountability.

H_{2}.

H_{o}: Effective e-Procurement implementation assessment does not significantly depend on public procurement governance.

H_{a}: Effective e-Procurement implementation assessment significantly depends on public procurement governance.

H_{3}.

H_{o}: Effective e-Procurement implementation assessment does not significantly depend on dematerialization.

H_{a}: Effective e-Procurement implementation assessment significantly depends on dematerialization.

H_{4}.

H_{o}: Effective e-Procurement implementation assessment does not significantly depend on legal, regulatory, and policy framework.

H_{a}: Effective e-Procurement implementation assessment significantly depends on legal, regulatory, and policy framework.

Variables in

Using

The essence of research design is to translate a research problem into data for analysis to impart pertinent answers to research questions at the least cost [

The population of the study is the PE officers and bidders of RHD. In this study, the population is known. Considering the known population, the total sample size of the survey is determined about 402 where PE officers, bidders, and FGD are 206, 145, and 51, respectively. KII, FGD, and survey questionnaire tools are adopted to conduct the research study. The primary data was collected from 4 wings, 11 zones, 31 circles, and 70 divisions of RHD in Bangladesh to maintain space triangulation.

Hypothesis tests were done on 4 dependent variables: Accountability, Procurement Governance, Dematerialization, Legal, Regulatory, and Policy Framework. Multiple Linear Regression (MLR) model using SPSS version 23 software was used for data analysis to test 4 hypotheses of the proposed conceptual framework. Each dependent variable had some influencing factors i.e. independent variables shown in

The survey used both structured and unstructured questionnaires keeping in line with the three objectives of the study. Data were collected from remote PE offices and bidders using email and telephone. Telephonic communication helped to clarify some answers to specific questions. The researchers collected the secondary data through document study and by reviewing the relevant literature.

The findings of gender issues indicated that 93.7% of PE officers were male while 6.3% of respondents were female. This finding implies that the majority of the PE officers in RHD are males.

The findings regarding the age of the respondents state that the age of PE officers started from 22 years and the maximum age is 59 years. The mean age is 41.21 years. The standard deviation of the PE officer’s age is 9.661 that indicates the average respondents are mid-level officers and they possess sufficient knowledge to be involved with the e-procurement systems.

The study about respondents’ education level is that principal PE officers i.e. 53.4% hold Bachelor of Science in engineering degrees. It is argued that significant respondents are educated who handle the e-procurement system in RHD.

The findings indicate that Sub Asst. Engineers, Asst. Engineers, Sub Divisional Engineers, Executive Engineers, Superintendent Engineers to senior Additional Chief Engineers are involved in giving opinions as of KII. It proves that data have been collected from all available posts in RHD who are actively engaged in the e-Procurement system at their offices in RHD.

From the findings, it is found that the PE officers’ e-GP roles are as PE users, TOC members, PE admin, and TEC members. Among them, a major 29.6% is PE users. These findings indicate that data was collected from the very relevant PE officers in RHD who had an active role in the e-Procurement system for public procurement. It also reveals that the credibility of the data is high.

In response to the question “how many times PE officers got e-GP training?” the findings indicate that 16.5% of PE officers did not participate in any e-GP related training. 44.2% of PE officers participated in e-GP related training 1 time, 29.1% of PE officers participated in e-GP related training 2 times, only 4.4% of PE officers participated in e-GP related training 3 times. Therefore, a significant proportion of PE officers had not attained at least 1 time of e-procurement training in RHD. This causes challenges and difficulties in handling e-Procurement related tender documents at the stage of the e-procurement process. The data suggest that PE officers training on e-procurement software at least once is a must to handle e-procurement software smartly.

The findings express that 100% of respondent bidders are male. The data reveals that there is no female bidder in RHD. It reflects the gender imbalance in bidders’ perception. The presence of a gender-equity policy will bring a balance between the male and female bidders.

The answer relating to the question of the highest educational degree of the bidders revealed that 62% of bidders possess a bachelor’s degree. Therefore, the major portion of the respondents has the potentiality to handle the e-procurement bidding process.

In response to the question “how many bidders got e-Procurement training?” indicates that 71% of bidders did not participate in any e-GP related training that was provided by CPTU. 26.2% of bidders participated in e-GP related training 1 time. Therefore, the major portion of bidders had not attained the necessary training on the implementation of the e-Procurement system in RHD. It creates challenges and difficulties in filling up the e-Procurement related tender documents at the stage of the bidding process. The data suggests that all registered bidders should be ensured to participate in the e-GP training.

One of the three objectives was “To predict significant effects of different factors that influence e-Procurement implementation”. The study tested four hypotheses H_{1} to H_{4}, for an effective e-Procurement implementation assessment model.

Hypothesis test on accountability H_{1}:

Let, dependent variable Y = accountability

And 4 independent variables are:

X_{1} = Collaboration between PE and bidders, X_{2} = PE officer’s behavioral changes,

X_{3} = Bidder’s behavioral changes, X_{4} = Value for money.

Regression Model-1 is: Y = β 0 + β 1 X 1 + β 2 X 2 + β 3 X 3 + β 4 X 4 + ϵ i

SPSS data interpretation:

The above model summary ^{2}.

The R-value represents the degree of correlation which is 0.556 (the “R” Column).The analyzed correlation value of 55.6% indicates a high degree (Positive strong) of correlation.

The R^{2} value (the “R Square” column) indicates how much of the total variation in the dependent variable “accountability”. By the R^{2} value of the above model summary

The regression model predicts the dependent variable accountability significantly well which reports that the regression equation fits the data well.

Here we got, sig = 0.000; i.e. p < 0.05

So, overall, the regression model is significant and statistically fit. Findings show that overall regression has a significant influence on increasing accountability by 4 independent variables.

Fitted regression model-1 is:

Y = 0.882 + 0.072 X 1 + 0.184 X 2 + 0.232 X 3 + 0.344 X 4 t 2.198 0.902 2.077 2.975 5.838 Sig ( 0.029 ) ( 0.368 ) ( 0.039 ) ( 0.003 ) ( 0.000 ) _{ }

Hypothesis test on procurement governance H_{2}—

Let, dependent variable Y = procurement governance

And the 6 independent variables are

X_{1} = capacity development, X_{2} = integrating principles, X_{3} = procurement planning publish, X_{4} = quality government tendering process, X_{5} = free from red-tapes, X_{6} = free from after contracts awarding

Regression Model-2 is Y = β 0 + β 1 X 1 + β 2 X 2 + β 3 X 3 + β 4 X 4 + β 5 X 5 + β 6 X 6 + ϵ i

SPSS data interpretation:

The above model summary ^{2}.

Model | R | R^{2} | Adjusted R^{2} | Std. Error of the Estimate | Change Statistics | ||||
---|---|---|---|---|---|---|---|---|---|

R^{2} Change | F Change | df1 | df2 | Sig. F Change | |||||

1 | 0.55^{a} | 0.309 | 0.295 | 0.568 | 0.309 | 22.454 | 4 | 201 | 0.000 |

^{a}Predictors: (Constant), Value for money, Collaboration between PE and bidders, bidders behavioral changes, PE officers behavioral changes.

Model | R | R^{2} | Adjusted R^{2} | Std. Error of the Estimate | Change Statistics | ||||
---|---|---|---|---|---|---|---|---|---|

R^{2}^{ } Change | F Change | df1 | df2 | Sig. F Change | |||||

1 | 0.883^{a} | 0.779 | 0.773 | 0.261 | 0.779 | 117.164 | 6 | 199 | 0.000 |

^{a}Predictors: (Constant), Free from after contracts awarding, Capacity development, Procurement planning publish, Quality government tendering process, Integrating principles, Free from red-tapism.

The R-value represents the degree of correlation, which is 0.883 (the “R” Column). The analyzed correlation value of 88.3% indicates a high degree (positive strong) of correlation.

The R^{2} value (the “R Square” column) indicates how much of the total variation in the dependent variable “procurement governance”. By the R^{2} value of the above model summary

The regression model predicts the dependent variable procurement governance significantly well which reports that the regression equation fits the data well (i.e., predicts the dependent variable).

Here we got, sig = 0.000; i.e. p < 0.05

So, overall, the regression model is significant and statistically fit. Analysis tells that overall regression has a significant influence on increasing procurement governance by 6 independent variables. In another word, the independent variables reliably predict the dependent variable.

Fitted regression model-2 is:

Y = 0.353 + 0.060 X 1 − 0.024 X 2 + 0.002 X 3 + 0.757 X 4 + 0.094 X 5 + 0.044 X 6 t 1.891 1.592 − 0.568 0.064 18.503 2.463 1.211 Sig ( 0.0 6 0 ) ( 0. 113 ) ( 0. 571 ) ( 0. 949 ) ( 0.000 ) ( 0.0 15 ) ( 0. 227 )

Hypothesis test on dematerialization H_{3}—

Let, dependent variable Y = dematerialization and the 2 independent variables are

X_{1} = paperless procurement process, X_{2} = reduced archiving cost

Regression Model-3 is Y = β 0 + β 1 X 1 + β 2 X 2 + ϵ i

SPSS data interpretation:

The above model summary ^{2}.

The R-value represents the degree of sample correlation which is 0.946 (the “R” Column). The analyzed correlation value is 94.6% which indicates a high degree (positive strong) of correlation.

The R^{2} value (the “R Square” column) indicates how much of the total variation in the dependent variable “dematerialization”. By the R^{2} value of the above

Model | R | R^{2} | Adjusted R^{2} | Std. Error of the Estimate | Change Statistics | ||||
---|---|---|---|---|---|---|---|---|---|

R^{2} Change | F Change | df1 | df2 | Sig. F Change | |||||

1 | 0.946^{a} | 0.894 | 0.893 | 0.233 | 0.894 | 859.672 | 2 | 203 | 0.000 |

^{a}Predictors: (Constant), Reduced archiving cost, Paperless procurement process.

model summary

The regression model predicts the dependent variable dematerialization significantly well which reports that the regression equation fits the data well (i.e., predicts the dependent variable).

Here we got, sig = 0.000; i.e. p < 0.05

So, overall, the regression model is significant and statistically fit. Analysis tells that overall regression has a significant influence on increasing dematerialization by 2 independent variables.

Fitted regression model-3 is:

Y = 0.205 + 0.894 X 1 + 0.065 X 2 t 2.209 29.533 2.244 Sig ( 0.028 ) ( 0.000 ) ( 0.026 )

Hypothesis test on legal, regulatory, and policy framework H_{4}—

Let, dependent variable Y = legal, regulatory, and policy framework and the 5 independent variables are

X_{1} = overall CPTU performance, X_{2} = PPA 2006 (and subsequent amendments), X_{3} = PPR 2008 (and subsequent amendments), X_{4} = e-GP guideline 2011, X_{5} = implementation of digital signature

Regression Model-4 is: Y = β 0 + β 1 X 1 + β 2 X 2 + β 3 X 3 + β 4 X 4 + β 5 X 5 + ϵ i

SPSS data interpretation:

The above model summary ^{2}.

The R-value represents the degree of sample correlation which is 0.913 (the “R” Column). The analyzed correlation value of 91.3% indicates a high degree (positive strong) of correlation.

The R^{2} value (the “R Square” column) indicates how much of the total variation in the dependent variable “legal, regulatory and policy framework” can be explained by the influence of the 5 independent variables. By the R^{2} value of the above model summary

The regression model predicts the dependent variable legal, regulatory, and policy framework significantly well fit which reports that the regression equation fits the data well (i.e., predicts the dependent variable).

Here we got, sig = 0.000; i.e. p < 0.05

So, overall, the regression model is significant and statistically fit. Analysis tells that overall regression has a significant influence on increasing legal, regulatory, and policy framework by 5 independent variables.

Fitted regression model-4:

Y = 0.390 + 0.013 X 1 + 0.022 X 2 + 0.047 X 3 + 0.846 X 4 + 0.006 X 5 t 2.539 0.427 0.411 1.022 24.718 0.469 Sig ( 0.0 12 ) ( 0. 67 0 ) ( 0. 66 0 ) ( 0. 3 0 8 ) ( 0.000 ) ( 0.0 64 )

This section presents the key points of the results of the previous subsection without statistical calculations. Key findings of each hypothesis are summarized in the following

R values of hypothesis H_{1} - H_{4} indicate that dependent variables accountability, procurement governance, dematerialization, (legal, regulatory, and policy frame-work) have a mutually strong positive correlation with its corresponding independent variables.

R^{2} values of hypothesis H_{1} - H_{4} indicate that dependent variables accountability, procurement governance, dematerialization, (legal, regulatory, and policy framework) have dependencies on their corresponding independent variables. Noted that this was an overall measure of the strength of association and did not reflect the extent to which particular independent variables are associated with the dependent variables.

So, overall, the regression model is highly significant and statistically fit.

The coefficient value of hypothesis H_{1} - H_{4} indicates that independent variables have a significant influence on the dependent variable.

Model | R | R^{2} | Adjusted R^{2} | Std. Error of the Estimate | Change Statistics | ||||
---|---|---|---|---|---|---|---|---|---|

R^{2} Change | F Change | df1 | df2 | Sig. F Change | |||||

1 | 0.913^{a} | 0.834 | 0.830 | 0.268 | 0.834 | 200.489 | 5 | 200 | 0.000 |

^{a}Predictors: (Constant), Digital signature implementation, Overall CPTU Performance, e-GP guideline 2011, PPR 2008 and amendments, PPA 2006 and amendments.

Hypothesis # | R | R^{2} | ANOVA Test | Alternative hypothesis H_{a} | Evidence from the test result |
---|---|---|---|---|---|

H_{1} | 0.556 | 0.309 | sig = 0.000; i.e. p < 0.05 | β_{1} ≠ 0, β_{2} ≠ 0, β_{3} ≠ 0, β_{4 }≠ 0 | Significant |

H_{2} | 0.883 | 0.779 | sig = 0.000; i.e. p < 0.05 | β_{1} ≠ 0, β_{2} ≠ 0, β_{3} ≠ 0, β_{4} ≠ 0, β_{5} ≠ 0, β_{6} ≠ 0 | Significant |

H_{3} | 0.946 | 0.894 | sig = 0.000; i.e. p < 0.05 | β_{1} ≠ 0, β_{2} ≠ 0 | Significant |

H_{4} | 0.913 | 0.834 | sig = 0.000; i.e. p < 0.05 | β_{1} ≠ 0, β_{2} ≠ 0, β_{3} ≠ 0, β_{4} ≠ 0, β_{5} ≠ 0 | Significant |

Source: Researcher’s field survey data analysis, 2020.

The test result of

This paper proposed a conceptual framework for e-Procurement assessment model based on the critical literature review and research gap analysis comprising 4 dependent variables like accountability, procurement governance, dematerialization, and (legal, regulatory, and policy framework) through FGD, KII, and survey questionnaires on RHD of the Government of Bangladesh. Hypothesis tests H_{1} - H_{4} using a regression model from the gathered data confirmed that there exists a strong correlation between each dependent variable with its corresponding independent variables. Test findings indicated that for 4 cases, H_{o} (null hypothesis) is rejected and Ha (alternative hypothesis) is accepted. The key message of this paper is that the proposed model for e-Procurement implementation assessment can be implemented to improve the e-GP policy guidelines.

The authors declare no conflicts of interest regarding the publication of this paper.

Rashid, Md.A. and Uddin, M.S. (2021) Analysis of a Conceptual Model and Assessment of the e-Procurement System in Bangladesh. Journal of Computer and Communications, 9, 64-76. https://doi.org/10.4236/jcc.2021.911005