Available via license: CC BY-SA 4.0
Content may be subject to copyright.
Jurnal Teknologi Informasi dan Pendidikan
Volume 16, No. 1, March 2023
https://doi.org/10.24036/tip.v16i1.720
P.ISSN: 2086 – 4981
E.ISSN: 2620 – 6390
tip.ppj.unp.ac.id
Task Technology Fit Adoption in the Recruitment Process Using
Google Form for IPSM Members
Muhamad Sigid Safarudin1*, Yunesman1, Hermansyah1
1Study Program Information Systems, STMIK Putera Batam, Batam, Indonesia
*Corresponding Author: muhamadsigidsafarudin@gmail.com
Article Information
ABSTRACT
Article history:
No. 720
Rec. Mei 10, 2023
Rev. July 31, 2023
Acc. September 16, 2023
Pub. September 24, 2023
Page. 156 – 173
In previous studies, it was found that there were
inconsistencies in research results related to the effect of TAC
on TTF, TEC on TTF, INC on TTF, TTF on USP, TTF on USS,
TTF on BEN, TTF on UTI, TTF on PEU, and TTF on PES. Many
companies/organizations use Google Forms for employee
recruitment, which requires many candidates. The purpose of
this study is to empirically examine the effect of Task
Characteristics, Technology Characteristics, and Individual
Characteristics on Task Technology Fit and its influence on
Individual Performance, User Satisfaction, Benefit, Utilization,
Perceived Ease of Use, and Perceived Usefulness in the
recruitment process using the google form. This research is
quantitative research that will examine the effect of the
variables, namely the independent variable and the dependent
variable, by testing the hypothesis. The population in this study
was 219 members of the Association of Human Resource
Practitioners (IPSM/ Ikatan Praktisi Sumber Daya Manusia).
The sample used in this study was 71 respondents who
processed data using Smart PLS 3.2.9. The results of this study
indicate that 8 of the nine hypotheses show a significant and
positive effect. This shows that IPSM members have used the
Google form to adopt Task Technology Fit in their recruitment
process. However, it is necessary to conduct further research
with more respondents.
This is an open access article under the CC BY-SA license.
Keywords:
▪ Adoption
▪ Task Technology Fit
▪ Google Form
▪ Recruitment Process
▪ Technology
Jurnal Teknologi Informasi dan Pendidikan
Volume 16, No. 1, March 2023
https://doi.org/10.24036/tip.v16i1.720
157
P.ISSN: 2086 – 4981
E.ISSN: 2620 – 6390
tip.ppj.unp.ac.id
1. INTRODUCTION
Task Characteristics (TAC) has an effect on the Task Technology Fit (TTF) [1]. Other
research also shows a significant relationship between Task Characteristics (TAC) and TTF
[2]. Whereas in research [3] showed the opposite results, namely that TAC had no effect on
TTF. Technology Characteristic (TEC) has a significant relationship to TTF [5]. In other
studies also shown that TEC has a positive effect on TTF [1]. Understanding Technology
Characteristics is a tool used by individuals to help complete their tasks. However, in its
application there is a discrepancy with the Technology Characteristics criteria [6]. Previous
studies have shown that Individual Characteristics (INC) always have a positive effect on
TTF. However, these results do not support the research of [7] and [8] which states that
individual characteristics do not affect TTF. TTF has no effect on User Satisfaction (USS).
Whereas in other studies it was explained that TTF had a significant positive effect on user
satisfaction [4]. The conclusion of the study [2] also shows that there is a significant
relationship between TTF which has an effect on USS. The use of technology should provide
many benefits for its users as in research [4]. Where users are free to access the system
wherever and whenever. This is in line with [9] which states that TTF has an effect on Benefit
(BEN). However, this is contradictory to research [10] where the technology that has offered
its users a level of acceptance is not as widespread as expected. It was explained that
Utilization of technology is very influential in an organization. TTF has a significant effect
on Utilization (UTI) [11]. Likewise in research [12] also showed the same results that there
is a positive effect of TTF on UTI. However, this study did not show consistency with
research conducted by [13] where the results showed that TTF had no significant positive
effect on UTI. Research [14] shows that in the third hypothesis, namely TTF on Perceived
Ease of Use (PEOU), it shows a positive and significant effect. This is also in line with the
results of research [15] which states that TTF has a positive and significant effect on PEOU.
Research [16] shows that a poor TTF is associated with a low ease-of-use score. PEOU is
defined as someone's belief that using the system will be free from effort [1]. Inconsistency
occurs in research [17] which concludes that PEOU has no effect on acceptance of TTF. And
TTF has a significant effect on Perceived Usefulness (PES) turns out to be unproven and
states a negative relationship [18]. Whereas in another study [19] in one of the third
hypotheses stated that the Ideal TTF would affect PES to produce evidence to the contrary,
namely a positive relationship.
Information technology supports the achievement of business objectives in the
organization [20]. The use of information technology is very helpful in activities that support
administrative processes[21]. Likewise, can be applied in the recruitment process. Many
companies use this method as part of administration system for reducing existing problem
and help to improve administration system service [22]. Therefore, many
companies/organizations use google forms for the recruitment process of employees who
Jurnal Teknologi Informasi dan Pendidikan
Volume 16, No. 1, March 2023
https://doi.org/10.24036/tip.v16i1.720
158
P.ISSN: 2086 – 4981
E.ISSN: 2620 – 6390
tip.ppj.unp.ac.id
need a large number of candidates. This trend has increased during the Covid19 pandemic
from 2019 until now.
Source: Batamindo
Figure 1. Companies in Batamindo Industrial Estate that use Google Forms in the recruitment
process
The data above shows that as many as 63% (46 of the total existing companies)
widely utilize and adopt the use of Google Forms in the employee recruitment process. So
this research was conducted to find and prove the consistency of the influence of previous
research, which still found differences in results, and to contribute to the development of
the Task Technology fit variable model related to the variable in Fit-Viability Theory (FVT),
namely User Performance (USP). Related to Expectation Confirmation Theory (ETC),
namely User Performance (USP), also related to Delone And Mclean IS Success Mode
(DIMS), namely Benefits (BEN), Utilizations (UTI), and finally, related Technology
Acceptance Model (TAM), namely Perceived Ease of Usefulness (PES), and Perceived Ease
of Use (PEU) in implementing recruitment using the Google Form.
2. RESEARCH METHOD
The research framework model developed in this study is where the Task
Technology Fit (TTF) model is added to the Perceived Usefulness (PES) variable, Perceived
Ease of Use (PEU) from the Technology Acceptance Model (TAM) model, User Performance
(USP) from the Fit- Viability Theory (FVT), User Satisfaction (USS) from the Expectation
Confirmation Theory (ETC) model, as well as Benefits (BEN) and Utilization (UTI) from the
Delone And Mclean IS Success Model (DIMS) model as in previous studies [23].
231
46
17
2
0
5
10
15
20
25
30
35
40
45
50
Total
Count of
MICROSOFT FORM
Count of AGENT
Count of JOB
PORTAL
Count of GOOGLE
FORM
Count of
REKRUITMENT
EMAIL
Jurnal Teknologi Informasi dan Pendidikan
Volume 16, No. 1, March 2023
https://doi.org/10.24036/tip.v16i1.720
159
P.ISSN: 2086 – 4981
E.ISSN: 2620 – 6390
tip.ppj.unp.ac.id
Figure 2. Framework Model
As the hypothesis for this research can be formulated as follow: H1. Task
Characteristics (TAC) influences Task Technology Fit (TTF)[1][2], H2. Technology
Characteristics (TEC) influences Task Technology Fit (TTF)[1][2], H3. Individual
Characteristic (INC) influences Task Technology Fit (TTF), H4. Task Technology Fit (TTF)
affects User Performance (USP)[3][24], H5. Task Technology Fit (TTF) has an effect on User
Satisfaction (USS)[25][18][4], H6. Task Technology Fit (TTF) affects Benefits (BEN)[9], H7.
Task Technology Fit (TTF) affects Utilization (UTI) [7][3], H8. Task Technology Fit (TTF)
influences Perceived Ease of Use (PEU)[26][27][28][29] and H9. Task Technology Fit (TTF)
has an effect on Perceived Usefulness (PES)[30][26][27]
2.1. Data Collection
The population in this study is members of the Association of Human Resources
Practitioners (IPSM/ Ikatan Praktisi Sumber Daya Manusia) registered in the WAG
(WhatsApp Group) amounting to 219 people. The smallest part of the population is referred
to as a sample. The sampling technique used is probability sampling it allows all members
of the population to get an equal chance of being selected for a sample. The sample selected
in this study was a member of the IPSM. The formula for determining the sample size used
is based on Slovin. Because IPSM members come from different companies and
organizations, getting all responses from these respondents will be very difficult. Therefore,
an error rate limitation of 10% is used. The data needed in this study is data on the Task
Technology Fit model added variables Perceived Usefulness (PES), Perceive Ease of Use
(PEU), User Performance (USP), User Satisfaction (USS), Benefits (BEN), and Utilization
(UTI). Meanwhile, data collection in this study was carried out in several ways providing
questionnaires to related parties, namely IPSM members in the WAG group.
2.2. Data measurement
Data measurement in this study uses the Likert scale [31] [32] where the Likert scale
is used to measure attitudes, opinions, and perceptions of a person or group of people about
social events or phenomena [33] The Likert scale used uses a range of 1 to 5 where 1 =
Jurnal Teknologi Informasi dan Pendidikan
Volume 16, No. 1, March 2023
https://doi.org/10.24036/tip.v16i1.720
160
P.ISSN: 2086 – 4981
E.ISSN: 2620 – 6390
tip.ppj.unp.ac.id
Strongly Disagree (STS), 2 = Disagree (TS), 3 = Undecided (RR), 4 = Agree (S), 5 = Strongly
Agree (SS). The instrument in this study is in the form of a questionnaire using a Google
form that has been arranged in such a way according to the variables to be studied. The
Google form link is as follows: https://forms.gle/L3zJ2PCcsTFs9qf37. The link is sent to each
IPSM member randomly through WA Broadcast. The software used in this study uses
Smart PLS (Partial Least Square) 3.2.9 In previous studies, many also used Smart PLS
software with different versions [35][36][37]. The steps of the analysis carried out are:
2.2.1. Designing a measurement model (outer model).
Conducted several tests, namely the first Convergent Validity Test is assessed based
on the value of the loading factor, known as outer loading. The convergent validity test
consists of Loading Factor / Outer Loading and Average Variance Extracted (AVE).
According to [38] the value of the loading factor must be more than 0.7 for confirmatory
research and between 0.6 – 0.7 for exploratory research. While the loading factor for
reflective model measurements is ≥ 0.708 [39]. The average variance inflation factor (AVE)
value must be greater than 0.5. The AVE is defined as the average variation of each
measurement item contained by a variable. How far the overall variable is can explain the
variation of measurement items where this measure also illustrates how well the convergent
validity of the variable [40].According to [39] the value (AVE) ≥ 0.50. Both Discriminant
Validity Tests This test is seen from the results of the Fornell-Lacker Criterion, Cross
Loadings, and Heterotraiit-Monotrait Ratio (HTMT). Cross Loading to measure indicators
of each construct/variable used. While Fornell-Lacker Criterion and HTMT measure
variable levels. The Fornell-Lacker Criterion shows a model has good discriminant validity
when the AVE root of the variable is greater than the correlation between variables. While
HTMT is the ratio of Heterotrait (average correlation between measurement items of
different variables) with the root of geometric multiplication Monotrait (correlation between
items measuring the same variable). With a recommended value below 0.85 or 0.90 [40]. The
third Reliability Test is carried out in two ways, namely with Cronbach's Alpha and
Composite Reliability (CR). To assess construct reliability the CR value must be greater than
0.70. However, the use of Cronbach's Alpha to test construct reliability will give a lower
value (underestimate) so it is more advisable to use Composite Reliability. Where according
to [40] the minimum CR value is 0.70. Fourth, the multicollinearity assumption test. This
examination can be seen from the VIF (Variance Inflated Factor). The limits used in this test
are usually expressed in VIF values at the indicator level > 5. So if the VIF value of the
indicator > 5, then there is a multicollinearity problem. Meanwhile, according to [39], there
is a collinearity problem if VIF ≥ 3-5. Therefore, the way to overcome it is one indicator that
has a strong correlation or eliminated
Jurnal Teknologi Informasi dan Pendidikan
Volume 16, No. 1, March 2023
https://doi.org/10.24036/tip.v16i1.720
161
P.ISSN: 2086 – 4981
E.ISSN: 2620 – 6390
tip.ppj.unp.ac.id
2.2.2. Designing a measurement model (outer model)
Conducted several tests, namely the first Convergent Validity Test is assessed based
on the value of the loading factor, known as outer loading. The convergent validity test
consists of Loading Factor / Outer Loading and Average Variance Extracted (AVE).
According to [38], the value of the loading factor must be more than 0.7 for confirmatory
research and between 0.6 – 0.7 for exploratory research. While the loading factor for
reflective model measurements is ≥ 0.708 [39]. The average variance inflation factor (AVE)
value must be greater than 0.5. The AVE is defined as the average variation of each
measurement item contained by a variable. How far the overall variable is can explain the
variation of measurement items where this measure also illustrates how well the convergent
validity of the variable [40]. According to [39] the value (AVE) ≥ 0.50. Both Discriminant
Validity Tests This test is seen from the results of the Fornell-Lacker Criterion, Cross
Loadings, and Heterotraiit-Monotrait Ratio (HTMT). Cross Loading to measure indicators
of each construct/variable used. While Fornell-Lacker Criterion and HTMT measure
variable levels. The Fornell-Lacker Criterion shows a model has good discriminant validity
when the AVE root of the variable is greater than the correlation between variables. While
HTMT is the ratio of Heterotrait (average correlation between measurement items of
different variables) with the root of geometric multiplication Monotrait (correlation between
items measuring the same variable). With a recommended value below 0.85 or 0.90 [40]. The
third Reliability Test is carried out in two ways, namely with Cronbach's Alpha and
Composite Reliability (CR). To assess construct reliability the CR value must be greater than
0.70. However, the use of Cronbach's Alpha to test construct reliability will give a lower
value (underestimate) so it is more advisable to use Composite Reliability. Where according
to [40] the minimum CR value is 0.70. Fourth, the multicollinearity assumption test. This
examination can be seen from the VIF (Variance Inflated Factor). The limits used in this test
are usually expressed in VIF values at the indicator level > 5. So if the VIF value of the
indicator > 5, then there is a multicollinearity problem. Meanwhile, according to Hair et al.,
(2019), there is a collinearity problem if VIF ≥ 3-5. Therefore, the way to overcome it is one
indicator that has a strong correlation or eliminated
2.2.3. Designing a structural model (inner model)
Where viewed from several sizes, namely: First R Square. According to Hair et al.
(2019), R square values are 0.75 (high), 0.50 (moderate), and 0.25 (low). Second, Q Square. In
[27], it is stated that the Q square is 0 (low), 0.25 (moderate), and 0.50 (high). Third F Square.
Where the F Square value is 0.02 (low), 0.15 (medium) 0.35 (large) [28]. The fourth goodness
of Fit index, where Yamin (2023) explained that the GoF (Goodness of Fit) values are 0.10
(low GoF), 0.25 (medium GoF), and 0.36 (high GoF), and the calculation is done manually
Jurnal Teknologi Informasi dan Pendidikan
Volume 16, No. 1, March 2023
https://doi.org/10.24036/tip.v16i1.720
162
P.ISSN: 2086 – 4981
E.ISSN: 2620 – 6390
tip.ppj.unp.ac.id
2.2.4. Fifth SRMR (Standardized Root Mean Square Residual)
As a reference, an SRMR value below 0.08 indicates a fit model, but another opinion
states that an SRMR of less than 0.10 is still acceptable [40]. Sixth PLS Predict, where the
parameters used are RMSE (Root Mean Squared Error) or MAE (Mean Absolute Error) and
Q square predictive [39]. Seventh, Robustness Check and (Linearity, Endogeneity, and
Heterogeneity). For the assessment criterion of the structural model (inner model), the
second is significance. The guidelines used (two-tailed) t-values are 1.65 (significance level
= 10%), 1.96 (significance level = 5%), and 2.58 (significance level = 1%) [38].
2.2.5. Conduct hypothesis testing and interpretation
The hypothesis is made by looking at the direct influence (path coefficients) and
indirect influence (total indirect effect) through the bootstrapping menu.
3. RESULTS AND DISCUSSION
The results of the analysis in this study are generally divided into three, namely the
results of demographic characteristics analysis, the results of descriptive analysis of
questionnaires, and finally the results of SEM analysis.
3.1 Demographic characteristics
Figure 3. Demographic characteristics result
Jurnal Teknologi Informasi dan Pendidikan
Volume 16, No. 1, March 2023
https://doi.org/10.24036/tip.v16i1.720
163
P.ISSN: 2086 – 4981
E.ISSN: 2620 – 6390
tip.ppj.unp.ac.id
Male respondents were 31 (42%), and female respondents were 42 (58%). With
respondents aged <20 years, as much as 2%, >50 years, as much as 1%, 21-30 years, as much
as 37%, 31-40 years, as much as 23%, and 41-50 years as much as 37%. The service period of
respondents <1 year as much as 1%, >20 years as much as 10%, 11-20 years as much as 16%,
1-2 years as much as 4%, and 3-5 years as much as 40% and 6-10 years as much as 29%. The
position of Coordinator is 1%, Executive is 18%, Manager is 15%, Officer is 21%, Senior
Supervisor is 1%, and Staff is 44%. Coming from the Admin department as much as 7%,
Finance as much as 1%, GA as much as 19%, HR as much as 66%, and Others as much as
7%. With Diploma education levels (D1-D3) as much as 3%, S1 as much as 88%, SMU / SMK
/ equivalent as much as 4%, and S2 as much as 5%. Based on married status, as much as 53%
and as much as 47% are unmarried. Working in foreign investment as much as 79%,
domestic investment much as 15%, and MSMEs as much as 6%. Those who work in
companies engaged in services and manufacturing 3%, services/services as much as 14%,
and manufacturing as much as 83%. Companies with <100 employees, as much as 18%, >
6000 people, as much as 3%; 1000 – 2000 people, as much as 18%; 3000 – 4000 people, as
much as 1%; and 500 – 1000 people, as much as 42%.
3.2 Descriptive Analysis of Questionnaires
The results of this questionnaire description analysis include Descriptive Variable
TAC, TEC, INC, TTF, PES, PEU, USP, USS, BEN, and UTI. On average, respondents gave
answers with a score of 4 (agree), namely with a percentage of 41%, and a score of 5 (strongly
agree) with a percentage of 30%. Because it is still below 50%, the TAC of respondents is at
the middle level. On average, respondents gave answers with a score of 4 (agree), namely
with a percentage of 45% and a score of 5 with a percentage of 49%. Because it is still below
50%, it can be said that the TEC of respondents is at the middle level. On average,
respondents gave answers with a score of 4, namely with a percentage of 47% and a score
of 5 with a percentage of 48%. Because it is still below 50%, the respondents' INC is at the
middle level. On average, respondents gave answers with a score of 4, namely with a
percentage of 40% and a score of 5 with a percentage of 55%. The respondent's TTF is very
high because it is still above 50%. On average, respondents gave answers with a score of 4,
namely with a percentage of 40% and a score of 5 with a percentage of 56%. Because it is
still above 50%, the respondent's PES is at a high level. On average, respondents gave
answers with a score of 4, namely with a percentage of 46% and a score of 5 with a
percentage of 51%. Because it is still above 50%, the respondent's PEU is at a high level. On
average, respondents gave answers with a score of 4, namely with a percentage of 47% and
a score of 5 with a percentage of 49%. Because it is still below 50%, the respondent's USP is
at the middle level. On average, respondents gave answers with a score of 4, namely with a
percentage of 39% and a score of 5 with a percentage of 55%. Because it is still above 50%,
the respondent's USS is at a high level. On average, respondents gave answers with a score
of 4, namely with a percentage of 43% and a score of 5 with a percentage of 52%. The
Jurnal Teknologi Informasi dan Pendidikan
Volume 16, No. 1, March 2023
https://doi.org/10.24036/tip.v16i1.720
164
P.ISSN: 2086 – 4981
E.ISSN: 2620 – 6390
tip.ppj.unp.ac.id
respondent's BEN are high because it is still above 50%. On average, respondents gave
answers with a score of 4, namely with a percentage of 35% and a score of 5 with a
percentage of 58%. The respondent's BEN are high because it is still above 50%.
3.3 SEM Analysis
3.3.1 Measurement Model Analysis (Outer Model)
Convergent Validity Test in the third testing stage, outer loadings between 0.40 and
0.70 were no longer found. The results of this evaluation conclude that the evaluation of the
measurement model from the aspect of convergent validity is fulfilled. So, all indicators and
constructs in the model have met the criteria of the Convergent Validity test. Discriminant
Validity Fornell-Larcker Criterion shows valid results because the value of the root AVE
(Fornell-Larcker Criteria) is greater than the correlation between latent variables. At the
same time, the results of Cross Loadings all aspects of discriminant validity at the
measurement item level have been met. The result for the HTMT value of variable pairs is
less than 0.90. This means that the variable has good discriminant validity. Next, Test
Reliability / Construct Reliability / Unidimensionality Model Composite Reliability (CR).
Figure 4. Results of Composite Reliability & Cronbach's Alpha Third Stage
Composite Reliability and Cronbach's Alpha all variables have been above 0.7. Thus,
it can be concluded that the four variables have reliable reliability because they meet the
criteria of the Composite Reliability test. Next, Test the Multicollinearity Assumption where
because of the discovery of the multicollinearity problem, the fourth stage of testing is
carried out again. This test is carried out by eliminating indicators that have a value of VIF>5,
namely: BEN10(5.832), BEN3(5.128), BEN4(5.601), BEN6(5.411), BEN7(6.009), BEN9(5.601),
INC6(5.817), PES1(6.684), PES2(5.361), PES3(8.113), PES8(5.362), PEU4(5.554), PEU6(6.194),
PEU8(5.230), TTF10(5.265), TTF14(6.663), TTF16(8.212), TTF17(6.801), TTF5 (6.131),
TTF8(6.270), USP4(5.926), USP6(5.078), USS2(5,102), USS3(5,701), USS4(6,359), USS5(6,791),
USS6(7,986), USS7(8,085), USS9(5,158), UTI1(5,675), UTI2(6,529), UTI3 (7,325), UTI4(5,985),
UTI5(5,337), UTI6(5,101), UTI8(7,199), UTI9(7,404).
Jurnal Teknologi Informasi dan Pendidikan
Volume 16, No. 1, March 2023
https://doi.org/10.24036/tip.v16i1.720
165
P.ISSN: 2086 – 4981
E.ISSN: 2620 – 6390
tip.ppj.unp.ac.id
3.3.2 Structural Model Instrument Data Analysis (Inner Model)
To determine the significance of the path coefficient of the t-test (critical ratio)
obtained from the bootstrapping process (sampling method), the signs of the path
coefficient must correspond to the theory of research hypothesis.
Figure 5. Inner Model Diagram of the Bootstrapping Process
Evaluating R Square, it can be explained that the magnitude of the variation in the
BEN described by TAC, TEC, INC and TTF is 61.8% (moderate effect). The magnitude
variation in the PEU variable described by TAC, TEC, INC and TTF is 32.7% (low effect).
The amount of PES variable variation described by TA, TEC, INC and TTF is 33.6% (low
effect). The magnitude of the variation in the TTF variable described by TAC, TEC, INC and
TTF is 61.9% (moderate effect). The magnitude of the USP variable variation described by
TAC, TEC, INC and TTF is 64.2% (moderate effect). The magnitude of the USS variable
variation described by TAC, TEC, INC and TTF is 64.8% (moderate effect). The magnitude
of the variation in the UTI variable described by TAC, TEC, INC and TTF is 35.1% (moderate
effect). As for effect Size F Square for INC on TTF is 0.087 (low category). F Square for TAC
on TTF is 0.023 (low category). F Square for TEC on TTF is 0.314 (moderate category). F
Square for TTF to BEN is 1.619 (large category). F Square for TTF on PEU is 0.487 (large
category). F Square for TTF on PES is 0.507 (large category). F Square for TTF on USP is 1.794
(large category). F Square for TTF on USS is 1.840 (large category). F Square for TTTF to UTI
is 0.540 (large category). The results of the SRMR of this research model are 0.082 < 0.10, so
it can be interpreted that the model-built matches empirical data. As for Q Square, they were
searched by blank folding. Q square redundancy for BEN 0.436> 0 and above (0.25) (medium
predictive reliability). The TTF variable can predict the BEN variable. Q square redundancy
for PEU 0.224 > 0 and below (0.25) (low predictive reliability). The TTF variable can predict
Jurnal Teknologi Informasi dan Pendidikan
Volume 16, No. 1, March 2023
https://doi.org/10.24036/tip.v16i1.720
166
P.ISSN: 2086 – 4981
E.ISSN: 2620 – 6390
tip.ppj.unp.ac.id
the PEU variable. Q square redundancy for PES 0.220> 0 and below (0.25) (low predictive
reliability). The TTF variable can predict the PES variable. Q square redundancy for TTF
0.376> 0 and above (0.25) (medium predictive reliability). INC, TAC and TEC variables can
predict the TTF variable. Q square redundancy for USP 0.393> 0 and above (0.25) (medium
predictive reliability). The TTF variable can predict the USP variable. Q square redundancy
for USS 0.478> 0 and above (0.25) (medium predictive reliability). The TTF variable can
predict the USS variable. Q square redundancy for UTI 0.230> 0 and above (0.25) (low
predictive reliability). The TTF variable can predict the variable UTI. Based on the
calculation results, GoF (0.539) is included in the high GoF category. This can be explained
based on GoF criteria with values of 0.10 (low GoF), 0.25 (medium GoF) and 0.36 (high GoF),
thus indicating that empirical data can explain measurement models with a high degree of
fit. For the overall PLS Predict results based on data processing, most (there are 234
measurements out of 256 and only 22 are high-value), RMSE and MAE values, PLS models
are lower than LM models (Linear Regression Models); hence the model has medium
predictive power.
3.4 Direct Influence (Path Coefficients)
Hypothesis 1 TAC affects TTF in the recruitment process using google forms. The
effect of TACon TTF is 0.135. With a P value of 0.168 >0.05. So the effect is not significant. At
the same time, T Statistics TAC to TTF is 1,379 < T table 2,000. So this hypo research is
accepted and accepted. Hypothesis 2 TEC affects TTF in the recruitment process using
Google Forms. The effect of TEC on TTF is 0.493. With a P value is 0.000<0.05. At the same
time, T Statistics The TEC of TTF is 3,941>T table 2,000. So this hypo research is accepted.
Hypothesis 3 INC affects TTF in the recruitment process using google forms. The effect of
INC on TTF is 0.255. With a P value is 0.023 <0.05. At the same time, T Statistics INC of TTF
is 2,281 < T table 2,000. So this hypothesis is accepted. Hypothesis 4 TTF affects USP in the
recruitment process using google forms. The effect of TTF on USP is 0.801. With a P value is
0.000 <0.05. While T Statistics TTF against USP is 12,706> T table 2,000. So this hypothesis is
accepted. Hypothesis 5 TTF affects USS in the recruitment process using google forms. The
effect of TTF on USS is 0.805. With a P value is 0.000 <0.05. While T Statistics TTF to USS is
11,502>T table 2,000. So this hypothesis is accepted. Hypothesis 6 TTF affects BEN in the
recruitment process using google forms. The effect of TTF on BEN is 0.786. With a P value
is 0.000 <0.05. While T Statistics TTF to BEN is 12,403 > T table 2,000. So this hypothesis is
accepted. Hypothesis 7 TTF affects UTI in the recruitment process using google forms. The
effect of TTF on UTI is 0.592. With a P value is 0.000 <0.05. While T Statistics TTF to UTI is
8,557 > T table 2,000. So this hypothesis is accepted. Hypothesis 8 TTF affects PES in the
recruitment process using google forms. The effect of TTF on PEU is 0.572. With a P value
is 0.000 <0.05. While T Statistics TTF against PEU is 6,792 > T table 2,000. So this hypothesis
is accepted. Hypothesis 9 TTF affects PES in the recruitment process using google forms.
Jurnal Teknologi Informasi dan Pendidikan
Volume 16, No. 1, March 2023
https://doi.org/10.24036/tip.v16i1.720
167
P.ISSN: 2086 – 4981
E.ISSN: 2620 – 6390
tip.ppj.unp.ac.id
The effect of TTF on PES is 0.580. With a P value is 0.000 <0.05. While T Statistics TAC to
TTF is 7,439 > T table 2,000. So this hypothesis is accepted.
3.5 Confident Interval Path Coefficient
The magnitude of the influence of INC on TTF in a 95% confidence interval is
between 0.103 to 0.460. This means that if there is a treatment or effort to increase INC, the
effect of increasing TTF can increase to 0.460. The effect of TAC on TTF in a 95% confidence
interval is between -0.027 to 0.296. This means that if there is a treatment or effort to increase
TAC, the effect of increasing TTF can increase to 0.296. The effect of TTF on BEN in a 95%
confidence interval is between 0.672 to 0.881. This means that if there is a treatment or effort
to increase TTF, the effect of increasing BEN can increase to 0.881. The magnitude of the
effect of TTF on PEU in a 95% confidence interval is between 0.579 to 0.435. This means that
if there is a treatment or effort to increase TTF, the effect of increasing PEU can increase to
0.435. The effect of TTF on PES in a 95% confidence interval is between 0.459 to 0.712. This
means that if there is a treatment or effort to increase TTF, the effect of increasing PES can
increase to 0.712. The effect of TTF on USP in a 95% confidence interval is between 0.682 to
0.887. This means that if there is a treatment or effort to increase TTF, the effect of increasing
USP can increase to 0.887. The magnitude of the effect of TTF on USS in a 95% confidence
interval is between 0.662 to 0.892. This means that if there is a treatment or effort to increase
TTF, the effect of increasing USS can increase to 0.892. The effect of TTF on UTI in a 95%
confidence interval is between 0.499 to 0.736. This means that if there is a treatment or effort
to increase UTI, the effect of increasing USS can increase to 0.736. The magnitude of the
influence of TEC on TTF in a 95% confidence interval is between 0.264 to 0.684. This means
that if there is a treatment or effort to improve TEC, the effect of increasing TTF can increase
to 0.684.
3.6 Total Indirect Effects / Mediation
TTF mediates the effect of INC on BEN with a mediation path coefficient of 0.201
and significant with T-count / T Statistics where t statistic (2.142 > 2.000) where P Values is
0.033<0.05. Task TTF mediates the effect of TAC on BEN with a mediation path coefficient
of 0.106 with no significance where T is calculated with t statistic (1.341 < 2.000) and P Values
is 0.180>0.05. TTF mediates the effect of TEC on BEN with a mediation path coefficient of
0.388 and significant with T-count / T Statistics with t statistic (3.924> 2.000) where P Values
are 0.000<0.05. TTF mediates the effect of INC on PEU with a mediation path coefficient of
0.146 and significant with T Statistics with t statistic (2.042 > 2.000) where P Values are
0.042<0.05. TTF mediates the effect of TAC on PEU with a mediation path coefficient of 0.077
with no significance where T is calculated / T Statistics with t statistic (1.351< 2.000) and P
Values is 0.177>0.05. TTF mediates the effect of TEC on PEU with a mediation path
coefficient of 0.282 and significant with T Statistics with t statistic (3.363 > 2.000) where P
Values are 0.001<0.05. TTF mediates the effect of INC on PES with a mediation path
Jurnal Teknologi Informasi dan Pendidikan
Volume 16, No. 1, March 2023
https://doi.org/10.24036/tip.v16i1.720
168
P.ISSN: 2086 – 4981
E.ISSN: 2620 – 6390
tip.ppj.unp.ac.id
coefficient of 0.148 and significant with T Statistics with t statistic (2.038 > 2.000) where P
Values are 0.042<0.05. mediates the effect of TAC on PES with a mediation path coefficient
of 0.078 with no significance where T Statistics (1.333< 2.000) and P Values are 0.183>0.05.
TTF mediates the effect of TEC on PES with a mediation path coefficient of 0.286 with
significance where T Statistics (3.395>2.000) and P Values are 0.001<0.05. TTF mediates the
effect of INC on USP with a mediation path coefficient of 0.205 with significance where T
Statistics (2.097>2.000) and P Values are 0.036<0.05. TTF mediates the effect of TAC on USP
with a mediation path coefficient of 0.108 with no significance where T Statistics 1.342<2.000)
and P Values 0.180>0.05. TTF mediates the effect of TEC on USP with a mediation path
coefficient of 0.395 with significance where T Statistics (4.003>2.000) and P Values are
0.000<0.05. TTF mediates the effect of INC on USS with a mediation path coefficient of 0.206
with significance where T Statistics (2.248>2.000) and P Values are 0.025<0.05. TTF mediates
the effect of TAC on USS with a mediation path coefficient of 0.108 with no significance
where T Statistics (1.339<2.000) and P Values are 0.181>0.05. TTF mediates the effect of TEC
on USS with a mediation path coefficient of 0.397 with significance where T Statistics
(3.711>2.000) and P Values are 0.000<0.05. TTF mediates the effect of INC on UTI with a
mediation path coefficient of 0.151 with significance where T Statistics (2.023>2.000) and P
Values are 0.044<0.05. TTF mediates the effect of TAC on UTI with a mediation path
coefficient of 0.080 with no significance where T Statistics (1.367<2.000) and P Values are
0.172>0.05. TTF mediates the effect of TEC on UTI with a mediation path coefficient of 0.292
with significance where T Statistics (3.356>2.000) and P Values are 0.001<0.05.
3.7 Confident Interval Specific Indirect Effects
Within a 95% confidence interval, TTF mediated the effect of INC on BEN between
0.077 and 0.372. Within a 95% confidence interval, TAC mediated the effect of TTF on BEN
between -0.020 to 0.240. Within a 95% confidence interval, TTF mediates the effect of TEC
on BEN between 0.215 to 0.530. Within a 95% confidence interval, TTF mediated the effect
of INC on PEU between 0.057 to 0.287. Within a 95% confidence interval, TTF mediated the
effect of TAC on PEU between -0.016 to 0.167. Within a 95% confidence interval, TTF
mediated the effect of TEC on PEU between 0.144 and 0.411. Within a 95% confidence
interval, TTF mediated the effect of INC on PES between 0.055 to 0.297. Within a 95%
confidence interval, TTF mediates the effect of TAC on PES between -0.015 to 0.176. Within
a 95% confidence interval, TTF mediates the effect of TEC on PES between 0.158 to 0.419.
Within a 95% confidence interval, TTF mediates the effect of INC on USP between 0.076 to
0.389. Within a 95% confidence interval, TTF mediates the effect of TAC on USP between -
0.021s.d. 0.243. Within a 95% confidence interval, TTF mediates the effect of TEC on USP
between 0.218 to 0.530. Within a 95% confidence interval, TTF mediates the effect of INC on
USS between 0.083 and 0.365. Within a 95% confidence interval, TTF mediates the effect of
TAC on USS between -0.021 to 0.248. Within a 95% confidence interval, TTF mediates the
effect of TEC on USS between 0.205 and 0.558. Within a 95% confidence interval, TTF
Jurnal Teknologi Informasi dan Pendidikan
Volume 16, No. 1, March 2023
https://doi.org/10.24036/tip.v16i1.720
169
P.ISSN: 2086 – 4981
E.ISSN: 2620 – 6390
tip.ppj.unp.ac.id
mediates the effect of INC on UTI between 0.060 - 0.311. Within a 95% confidence interval,
TTF mediates the effect of TAC on UTI between -0.016 to 0.179. Within a 95% confidence
interval, TTF mediates the effect of TEC on UTI between 0.156 and 0.434
4. CONCLUSION
It can be concluded that the direct influence (path coefficients) of TAC, TEC and INC
on TTF has a positive influence. While TTF on USP, USS, BEN, UTI, PEU, and PES also has
a positive influence. Nevertheless, not all hypotheses are accepted. Of the nine hypotheses,
one hypothesis was rejected, namely the effect of TAC on TTF, which had an original sample
value of 0.135. The influence of the two variables is considered insignificant because the
statistic T value is smaller than the table T (1.379 < 2.000). Based on indirect effects (total
indirect effects) shows that TTF mediates the influence between other variables significantly
except to mediate the effect of TAC on BEN, mediate the effect of TAC on PEU, mediate the
influence of TAC on PES, mediate the influence of TAC on USP, mediating the effect of TAC
on USS and mediating the effect of TAC on UTI produces insignificant results. The results
of this study show that IPSM member organizations or companies have adopted Task
Technology Fit in their recruitment process using google forms. This is shown by the results
of the significant direct influence of all variables used on Task Technology Fit. Only one
variable is TAC, although it has a positive influence but not significant. It is recommended
that further research related to TTF related to the use of google forms by involving a larger
number of respondents. And also, research related to TAC that has an insignificant influence
is needed to be investigated further, as well as TAC with TTF mediation on variable BEN,
PEU, PES, USP, USS, and UTI is not significant. It is also necessary to conduct further
research involving other mediating variables to obtain consistent results.
REFERENCES
[1] Muhammad Faisaal Syahlan, “Penerimaan Pengguna Aplikasi Trafi Menggunakan
Model Unified Theory Of Acceptance And Use Of Technology (UTAUT) Dan Task
Technology Fit (TTF),” vol. 3, no. March, p. 6, 2021, [Online]. Available:
https://repository.uinjkt.ac.id/dspace/handle/123456789/57641
[2] Q. Al-Maatouk, M. S. Othman, A. Aldraiweesh, U. Alturki, W. M. Al-Rahmi, and A. A.
Aljeraiwi, “Task-technology fit and technology acceptance model application to
structure and evaluate the adoption of social media in academia,” IEEE Access, vol. 8,
pp. 78427–78440, 2020, doi: 10.1109/ACCESS.2020.2990420.
[3] N. J. Al-Khafaji, R. M. Abdullah, and M. A. Kashmoola, “Evaluating the tracking e-files
prototype project based on the task technology fit model,” Indones. J. Electr. Eng.
Comput. Sci., vol. 17, no. 2, pp. 728–738, 2019, doi: 10.11591/ijeecs.v17.i2.pp728-738.
[4] L. Wan, S. Xie, and A. Shu, “Toward an Understanding of University Students’
Continued Intention to Use MOOCs: When UTAUT Model Meets TTF Model,” SAGE
Jurnal Teknologi Informasi dan Pendidikan
Volume 16, No. 1, March 2023
https://doi.org/10.24036/tip.v16i1.720
170
P.ISSN: 2086 – 4981
E.ISSN: 2620 – 6390
tip.ppj.unp.ac.id
Open, vol. 10, no. 3, 2020, doi: 10.1177/2158244020941858.
[5] R. Bisma and G. S. Palupi, “Analisis Kesesuaian Tugas Teknologi Pembelajaran Daring
Terhadap Lingkungan Sosial pada Masa Pandemi Covid-19,” J. Emerg. Inf. Syst. …, vol.
02, no. 02, pp. 10–16, 2021, [Online]. Available:
https://ejournal.unesa.ac.id/index.php/JEISBI/article/view/38979%0Ahttps://ejournal.u
nesa.ac.id/index.php/JEISBI/article/download/38979/38744
[6] I. De Crystal, S. Farlinda, N. Nuraini, and A. P. Wicaksono, “Evaluasi Implementasi
Aplikasi Primary Care (P-Care) dengan Menggunakan Metode Task Technology Fit di
Puskesmas Patrang Kabupaten Jember Tahun 2019,” J-REMI J. Rekam Med. dan Inf.
Kesehat., vol. 1, no. 4, pp. 502–510, 2020, doi: 10.25047/j-remi.v1i4.2127.
[7] Mariana Anselia Gama, S. Bambang, and Fidiana, “Pengaruh Task-Technology Fit
Terhadap Prestasi Belajar Mahasiswa Akuntansi Dimediasi Oleh Pemanfaatan
Smartphone Suwardi Bambang Fidiana Sekolah Tinggi Ilmu Ekonomi Indonesia
(STIESIA) Surabaya,” 1995.
[8] K. Childa Maulina, Endang Siti Astuti, “Pengaruh Karakteristik Tugas , Teknologi
Informasi Dan Individu Terhadap TTF, Utilisasi Dan Kinerja,” JISIP Univ. Tribuwana
Tunggadewi, vol. 4, no. 1, pp. 108–119, 2015.
[9] E. R. Bravo and H. A. Bayona, “Explaining multidimensional Facebook benefits: A task-
technology fit approach,” in Proceedings of the Annual Hawaii International Conference on
System Sciences, 2020, vol. 2020-Janua, pp. 4474–4482. doi: 10.24251/hicss.2020.547.
[10] S. Naz, A. Akbar, P. Poulova, J. M. Martins, S. A. Haider, and M. Nuno, “Extending the
Role of Technology Continuance Theory and Task Technology Fit with Pre-Technology
Adoption and Post-Technology Adoption Behaviors,” no. September, pp. 1–23, 2021,
doi: 10.20944/preprints202109.0219.v1.
[11] D. Wahyuningsih, I. Noor, M. R. K. Muluk, F. I. Administrasi, and U. Brawijaya,
“Pengaruh Task-Technology Fit Dan Utilization Siakad Dan Ub-Feeder Terhadap
Dampak Kinerja Individu Di,” no. 2, pp. 30–48, 2019.
[12] S. A. Kurniabudi, “Jurnal Nasional Teknologi dan Sistem Informasi Model Persepsi
Penggunaan Media Sosial pada Perkuliahan dengan Modifikasi Task Technology Fit
dan Expectation Confirmation Theory,” J. Nas. Teknol. dan Sist. Inf., vol. 03, no. 25
Desember 2018, pp. 107–114, 2018, [Online]. Available:
http://teknosi.fti.unand.ac.id/%0AJurnal
[13] Z. Firdausi, R. Umilasari, and D. Arifianto, “Analisis Kesesuian Utilisasi Dan Kinerja E-
Learning Menggunakan Metode Task Technology Fit (Ttf) Terhadap Dosen Universitas
Muhammadiyah Jember,” no. 1510651137, 2019, [Online]. Available:
http://repository.unmuhjember.ac.id/id/eprint/3628
[14] I. Sulistyaningsih and J. Nugraha, “Analisis Penerimaan Pengguna Platform
Pembelajaran Virtual Learning Unesa (Vinesa) Menggunakan Task Technology Fit
(TTF) Dan Technology Acceptance Model (TAM) Di Masa Pandemi COVID-19,” J.
Pendidik. Adm. Perkantoran, vol. 10, no. 1, pp. 107–123, 2022, doi:
Jurnal Teknologi Informasi dan Pendidikan
Volume 16, No. 1, March 2023
https://doi.org/10.24036/tip.v16i1.720
171
P.ISSN: 2086 – 4981
E.ISSN: 2620 – 6390
tip.ppj.unp.ac.id
10.26740/jpap.v10n1.p107-123.
[15] F. S. Pratiwi, Evaluasi teknologi pembelajaran digital dan pengaruhnya terhadap niat
mahasiswa untuk menggunakan course pada zedemy, vol. X. 2021. [Online]. Available:
https://repository.its.ac.id/89020/
[16] O. Isaac, Z. Abdullah, A. H. Aldholay, and A. Abdulbaqi Ameen, “Antecedents and
outcomes of internet usage within organisations in Yemen: An extension of the Unified
Theory of Acceptance and Use of Technology (UTAUT) model,” Asia Pacific Manag.
Rev., vol. 24, no. 4, pp. 335–354, 2019, doi: 10.1016/j.apmrv.2018.12.003.
[17] C. Tesavrita and D. R. Dharsono, “Analisis Model TAM / TTF sebagai Dasar Usulan
Perbaikan untuk Pelaku Bisnis E-Commerce,” Semin. Appl. Res. Ind. Technol. SMART
Yogyakarta, 2009.
[18] K. B. Al-Gharbawi, “Task-Technology Fit of MIS and its Impact on MIS User
Acceptance and Satisfaction at UNRWA Relief and Social Services Area Offices - Gaza,”
Gaza, 2016.
[19] M. C. Hung, P. C. Talley, K. M. Kuo, and M. L. Chiu, “Exploring cloud-based bookstore
continuance from a deconstructed task–technology fit perspective,” J. Theor. Appl.
Electron. Commer. Res., vol. 16, no. 3, pp. 356–376, 2021, doi: 10.3390/jtaer16030023.
[20] A. Subhan, “Teknologi Informasi Dan Pendidikan Islam,” ADDABANA J. Pendidik.
Agama Islam, vol. 2, no. 2, pp. 1–12, 2019, doi: 10.47732/adb.v2i2.117.
[21] T. Rahayu, N. Matondang, and B. Hananto, “Audit Sistem Informasi Akademik
Menggunakan Metode Cobit 5,” J. Teknol. Inf. dan Pendidik., vol. 13, no. 1, pp. 117–123,
2020, doi: 10.24036/tip.v13i1.305.
[22] R. Marta, “Design and Build a Web-Based Public Service Administration System Using
the CodeIgniter Framework,” vol. 15, no. 2, 2023, doi: https://doi.org/10.24036/tip.v15i2.
[23] A. Gormantara, “Evaluation of the Success of the Academic Information System (
SIAMIK ) with the DeLone and McLean Model,” vol. 15, no. 2, 2023, doi:
https://doi.org/10.24036/tip.v15i2.
[24] J. R. Jardina, B. S. Chaparro, and S. Abdinnour, “Extending the Task-Technology Fit
(TTF) model to e-textbook usage by students and instructors,” Int. J. Inf. Commun.
Technol. Educ., vol. 17, no. 1, pp. 120–137, 2021, doi: 10.4018/IJICTE.2021010108.
[25] O. Isaac, Z. Abdullah, T. Ramayah, and A. M. Mutahar, “Factors determining user
satisfaction of internet usage among public sector employees in Yemen,” Int. J. Technol.
Learn. Innov. Dev., vol. 10, no. 1, pp. 37–68, 2018, doi: 10.1504/IJTLID.2018.091800.
[26] A. Y. Yaakop, Y. P. Shi, B. Foster, and J. Saputra, “Investigating e-wallet adoption of
COVID19 intra-period among Malaysian youths’: Integrated task-technology fit and
technology acceptance model framework,” Int. J. Data Netw. Sci., vol. 5, no. 3, pp. 295–
302, 2021, doi: 10.5267/j.ijdns.2021.6.004.
[27] F. Aligarh and Wijianto, “Analisis Sistem Informasi Debitur Dengan Technological
Acceptance Model Dan Task-Technology Fit,” Semin. Nas. dan Call Pap. Fak. Ekon. Univ.
Muhammadiyah Ponorogo, pp. 156–166, 2018.
Jurnal Teknologi Informasi dan Pendidikan
Volume 16, No. 1, March 2023
https://doi.org/10.24036/tip.v16i1.720
172
P.ISSN: 2086 – 4981
E.ISSN: 2620 – 6390
tip.ppj.unp.ac.id
[28] R. S. Rai and F. Selnes, “Conceptualizing task-technology fit and the effect on adoption
– A case study of a digital textbook service,” Inf. Manag., vol. 56, no. 8, p. 103161, 2019,
doi: 10.1016/j.im.2019.04.004.
[29] I. B. G. A. Permana and D. P. Setianto, “The Impact of Task Technology Fit, System
Quality and Information Quality To User Performance: Perceived Ease Usefullness and
Perceived Ease of Use As Mediation,” Int. J. Organ. Innov., vol. 12, no. 2, pp. 335–340,
2018, [Online]. Available: http://www.ijoi-online.orghttp//www.ijoi-online.org
[30] G. B. Akrong, Y. Shao, and E. Owusu, “Evaluation of organizational climate factors on
tax administration enterprise resource planning (ERP) system,” Heliyon, vol. 8, no. 6, p.
e09642, 2022, doi: 10.1016/j.heliyon.2022.e09642.
[31] E. F. Effendi and E. Besra, “Analisis Keterkaitan Store Image, Customer Satisfaction Dan
Repurchase Intention (Survei Pada Pelanggan Sjs Plaza),” J. Ilm. Mhs. Ekon. Manaj., vol.
4, no. No.1, pp. 372–384, 2019.
[32] D. Rosa Indah, A. Permata Putra, and M. Afriyan Firdaus, “Analysis of User Acceptance
Using UTAUT2 Model in KAI Access Application,” J. Teknol. Inf. dan Pendidik., vol. 15,
no. 2, 2022, [Online]. Available: https://doi.org/10.24036/tip.v15i2
[33] A. Andini and D. Surya, “Analisis Pengaruh Kualitas Produk Dan Citra Merek
Terhadap Keputusan Pembelian Dengan Word Of Mouth Sebagai Variabel
Intervening,” Sekol. Tinggi Ilmu Ekon. Indones., pp. 1–21, 2020, [Online]. Available:
http://repository.stei.ac.id/id/eprint/2420
[34] W. W. Winarno and A. MAFIS, “Faktor-faktor yang Mempengaruhi Penerimaan
Layanan E-Government di Kabupaten Gunungkidul,” 2020.
[35] A. Frida, “The Effect Of Pre Project Selling Strategy, Location And Quality Of Service
On Purchase Decisions And Trust As Intervening Variables On Housing Business In
The Satellite City Region Of Surabaya (Gresik And Sidoarjo),” Universitas 17 Agustus
Surabaya, 2020. [Online]. Available: http://repository.untag-sby.ac.id/12519/
[36] R. Setyadi and M. H. Baqi, “Analysis of the Use of the Bebunge Application Using End
User Computing Satisfaction Model,” J. Teknol. Inf. dan Pendidik., vol. 14, no. 1, pp. 83–
87, 2021, doi: 10.24036/tip.v14i2.423.
[37] R. Darni, L. Mursyida, and A. D. Samala, “Career Exploration System (C-EXSYS) in Era
Society 5.0 Based on Expert System,” J. Teknol. Inf. dan Pendidik., vol. 14, no. 2, pp. 131–
143, 2021, doi: 10.24036/tip.v14i2.491.
[38] M. . Rahmad Solling Hamid, S.E., M.M dan Dr. Suhardi M Anwar, Drs., Structural
Equation Modeling (SEM) Berbasis Varian : Konsep Dasar dan Aplikasi dengan Program
SmrtPLS 3.2.8 dalam Riset Bisnis, 1st ed. Jakarta Pusat: PT Inkubator Penulis Indonesia,
2019. [Online]. Available: https://www.ptonline.com/articles/how-to-get-better-mfi-
results
[39] J. F. Hair, J. J. Risher, M. Sarstedt, and C. M. Ringle, “When to use and how to report
the results of PLS-SEM,” Eur. Bus. Rev., vol. 31, no. 1, pp. 2–24, 2019, doi: 10.1108/EBR-
11-2018-0203.
