ORIGINAL RESEARCH

published: 14 July 2021

doi: 10.3389/fpsyg.2021.697306

The Basic Psychological Need

Satisfaction and Need Frustration at
Work Scale: A Validation Study
Anja Hagen Olafsen*, Hallgeir Halvari and Claus Wiemann Frølund

School of Business, University of South-Eastern Norway, Hønefoss, Norway

The aim of the present study was to adapt and validate the Basic Psychological
Need Satisfaction and Frustration Scale within self-determination theory (SDT) within
the work domain. Confirmatory factor analyses of three Norwegian samples and one
English sample as well as multi-group analyses to examine measurement invariance were
performed. The results showed that the adapted work-related scale with its six-factor
structure fitted the data well in all four samples, and partial measurement invariance
was obtained across samples and languages. Furthermore, internal consistencies for
the subscales were acceptable and the subscales predicted work-related correlates as
expected, demonstrating the criterion validity of the scale. The current study contributes
Edited by: to a unifying measurement for future research on one of the central underpinnings of SDT
Antonio Granero-Gallegos,
University of Almeria, Spain within the work domain.
Reviewed by: Keywords: basic psychological needs, self-determination theory, need satisfaction, need frustration, autonomy,
Rafael Burgueno, competence, relatedness, measurement
Universidad Isabel I de Castilla, Spain
Patrício Soares Costa,
University of Minho, Portugal INTRODUCTION
*Correspondence:
Anja Hagen Olafsen One of the most prominent motivational theories of our time is self-determination theory (SDT;
[email protected] Deci and Ryan, 2000; Ryan and Deci, 2017), which posits that humans have inherent basic
psychological needs that are the nutriments to human motivation, growth, flourishing, and well-
Specialty section: being. In particular, basic psychological needs theory is one of six mini-theories within SDT that
This article was submitted to theorizes that when humans’ basic psychological needs are satisfied humans thrive, while when
Organizational Psychology, these basic psychological needs are frustrated, maladjustment and even psychopathology can be
a section of the journal
the result (Vansteenkiste and Ryan, 2013; Vansteenkiste et al., 2020). Basic psychological needs are,
Frontiers in Psychology
hence, an essential part of the growing body of SDT research across various life domains, including
Received: 19 April 2021 that of work. Indeed, promoting need satisfaction (rather than need frustration) has been shown to
Accepted: 11 June 2021
be of essence for fostering autonomous work motivation, positive work attitudes and behaviors, as
Published: 14 July 2021
well as employee well-being (Deci et al., 2017; Olafsen and Deci, 2020).
Citation:
A unified understanding and approach to the basic psychological needs is essential to accumulate
Olafsen AH, Halvari H and Frølund CW
(2021) The Basic Psychological Need
research on this important mechanism within motivational research based on SDT. A measurement
Satisfaction and Need Frustration at scale for basic psychological need satisfaction and frustration developed by Chen et al. (2015)
Work Scale: A Validation Study. has been an important piece in these efforts, and the Basic Psychological Need Satisfaction and
Front. Psychol. 12:697306. Frustration Scale (BPNSFS) has become the go-to scale for assessing the mechanism of basic
doi: 10.3389/fpsyg.2021.697306 psychological needs within SDT. Currently, the scale has been adapted to various contexts,

Frontiers in Psychology | www.frontiersin.org 1 July 2021 | Volume 12 | Article 697306

Olafsen et al. Basic Psychological Needs Work Scale

cultures, and languages. However, to the authors’ knowledge, need frustration but where the presence of need frustration
there is still no formal validation of this measurement scale in the denotes the absence of need satisfaction (Vansteenkiste and
work domain. Given the increased use of SDT in organizational Ryan, 2013). Following this, as an active threat of the basic
psychology (Gagné and Deci, 2014), this scale would be fruitful psychological needs, need frustration has been shown to be an
to employ in future work. To have a common operationalization important predictor of detrimental consequences over and above
of basic psychological need satisfaction and frustration at work satisfaction of these needs (Bartholomew et al., 2011a; Chen et al.,
would enable the comparison of results both within and between 2015; Martinent et al., 2015).
contexts in the work domain. The purpose of the current study is
to adapt the BPNSFS into a work domain measure and validate
it in Norwegian and English. This contributes to (1) validation of Basic Psychological Needs in the
the scale to fit the work context and (2) validation of the scale in Workplace
Norwegian and English. Research building on SDT within organizational psychology
has gained increased attention in the last 10–15 years. This
Self-determination Theory body of research has addressed the links between motivation
SDT (Deci and Ryan, 2000; Ryan and Deci, 2017) is a and the dual concerns of performance and employee well-
macro theory of human motivation that has developed through being in organizations (Deci et al., 2017, Olafsen and Deci,
empirical work for the last six decades. Today, SDT represents 2020). A core element in much of this research is hence
a powerful framework of mini-theories that taps into different the support and satisfaction of the basic psychological needs
questions of human motivation and is used in a wide range of for autonomy, competence, and relatedness for promoting
different contexts such as sports, education, health, and work. quality work motivation, positive work functioning, and well-
A central underpinning of SDT is the basic psychological needs. being. Accordingly, these needs have been used to study both
Specifically, within SDT it is acknowledged that all humans have antecedents and outcomes of motivational processes at work.
a set of basic psychological needs that have to be satisfied for Based on this body of literature, a recent meta-analysis of
people to grow, flourish, and be physically and psychologically 99 studies with 119 distinct samples examined the accumulated
well. In contrast, if these needs are not satisfied, it will have antecedents and consequences of basic psychological need
physiological and psychological costs. Within SDT, three such satisfaction at work (Van den Broeck et al., 2016). The
basic psychological needs are identified through empirical work. study showed that various social-contextual factors have been
First, the need for autonomy (deCharms, 1968) refers to the examined in relation to the basic psychological needs, such
feeling of choice and concurrence with one’s actions. Second, the as job characteristics (various job resources and job demands)
need for competence (White, 1959), refers to the feeling of being and organizational contexts (manager support, organizational
effective and capable. Third, the need for relatedness (Baumeister support, justice, etc.). Furthermore, the meta-analysis pointed
and Leary, 1995) refers to the feeling of a connection to, caring to various outcomes of well-being and job attitudes that
for, and being cared for by other individuals and groups. have been the focus of previous studies. This goes to
Decades of research point to the benefits of need satisfaction show the importance of the basic psychological needs for
for quality of motivation, growth, functioning, and physical autonomy, competence, and relatedness as a central concept
and psychological wellness across life domains (Ryan and Deci, in understanding the motivational process going on in the
2017; Vansteenkiste et al., 2020). Moreover, in recent years, the workplace. In particular, basic psychological needs represent the
concept of need frustration has gained increased attention, where psychological mechanisms that explain how and why social-
research has shown detrimental consequences of getting the basic contextual factors in the workplace are related to quality of
psychological needs for autonomy, competence, and relatedness motivation and various important work outcomes.
frustrated (Bartholomew et al., 2011b; Chen et al., 2015; Olafsen As for need frustration, the same meta-analysis concluded
et al., 2017). Importantly, need frustration is emphasized as a that this concept needs further research because it has not
separate mechanism that leads to these outcomes, rather than received much attention in the literature so far. However, some
being at the opposite pole of a need satisfaction continuum studies with a focus on basic psychological need frustration do
(Bartholomew et al., 2011b; Cordeiro et al., 2016). In particular, exist. For instance, in a longitudinal study by Olafsen et al.
need frustration can be experienced when the basic psychological (2017), the experience of need frustration at work was associated
needs are actively undermined as a result of social contextual with higher levels of work-related stress, which predicted
influences (Vansteenkiste et al., 2020)—autonomy when one is higher levels of somatic symptoms, emotional exhaustion, and
forced to undertake a certain task in a certain way, competence absence due to sickness. Furthermore, Trépanier et al. (2015),
when one is told one cannot, or relatedness when one is being showed how need frustration, as predicted by job demands
excluded or rejected. In this way, need frustration is something (positive) and job resources (negative), related positively to
other than lack of need satisfaction, which represents a more controlled work motivation, which, in turn, related positively
passive obstruction of the basic psychological needs (e.g., not to psychological distress and psychosomatic complaints, and
having choices, lacking skills, not sharing similar interest with a negatively to work engagement and work performance. As
particular group). Need satisfaction and frustration is theorized the last example, although other studies do exist, Van den
to stand in an asymmetrical relation, where the absence of Broeck et al. (2014) found a mediating role of frustration
need satisfaction does not necessarily imply the presence of of the three basic psychological needs between job insecurity

Frontiers in Psychology | www.frontiersin.org 2 July 2021 | Volume 12 | Article 697306

Olafsen et al. Basic Psychological Needs Work Scale

and counterproductive work behavior. While these findings In sum, there is a need for a more consistent measure to
show important results for the role of basic psychological need assess both satisfaction and frustration of the basic psychological
frustration at work, more research is still needed in this area. needs within SDT in the work context. Recently, Longo et al.
This requires adequate tools for assessing need frustration in (2016) developed a scale assessing both need satisfaction and
combination with need satisfaction that can be used across need frustration, hence tackling some of the issue with previous
studies in the work domain. scales. This scale was validated in both the educational domain
and in the work domain, with the latter being validated in one
sample of MTurkers. Furthermore, Chen et al. (2015) developed a
Measurement of Basic Psychological general scale to assess need satisfaction and need frustration—the
Needs in the Work Context BPNSFS. The BPNSFS has quickly become the scale of preference
Several measures have been developed and used to assess the for the international community of SDT, with a lot of efforts being
basic psychological needs at work. For basic psychological need put into adapting it and validating it for different domains and in
satisfaction, published work has typically used the Basic Need different languages. Indeed, these efforts are promising in terms
Satisfaction at Work scale, a 21-item questionnaire (Deci et al., of having a valid and reliable measure that works across cultures
2001; Baard et al., 2004). This scale has been criticized for and life domains, including that of work. However, while the
reasons related to lack of content validity, formal validation scale has been used to capture this concept in the work domain
processes, and problems with reliability and intercorrelations (Schultz et al., 2015), it has not yet been rigorously validated
for the subscales. More recent work has employed the Work- in this domain. Based on the support and attention given to
Related Basic Need Satisfaction Scale developed by Van den the BPNSFS by the scholarly community, we believe we would
Broeck et al. (2010). While the Van den Broeck et al. (2010) be well-served by adding to this stream of studies by formally
scale has been argued to provide a better operationalization of validating it within the work domain.
basic psychological need satisfaction in the workplace because
it followed traditional validation processes, past research has The Present Study
noted problems with this scale as well (Olafsen et al., 2015; This study sought to validate the Norwegian and English
Olafsen and Halvari, 2017; Tafvelin and Stenling, 2018). In version of the BPNSFS adapted to the work domain in four
particular, with the introduction of need frustration, it can steps. (1) The reliability and factorial validity of the Basic
be argued that some of the reversed score items might tap Psychological Need Satisfaction and Frustration at Work Scale
into need frustration rather than need satisfaction. As already (BPNSFWS) were examined in three Norwegian samples and
mentioned, it has been acknowledged that these two concepts one English sample. A three-factor model differentiating between
are not merely at the opposite ends of a continuum, rather items for each of the three basic psychological needs (both
being distinct concepts related to somewhat different antecedents satisfaction and frustration items) was compared to a six-
and outcomes, and thus accounting for different motivational factor model differentiating between need satisfaction factors
processes (Bartholomew et al., 2011a). and need frustration factors [according to Chen et al. (2015)]
As a more recent concept within SDT, basic psychological using confirmatory factor analysis (CFA). (2), Measurement
need frustration in the work domain has so far mainly been invariance (MI) across the samples using multi-group CFA,
assessed with an adapted version of the first measurement scale testing for configural, metric, scalar, and residual invariance
for need frustration developed by Bartholomew et al. (2011a), (Putnick and Bornstein, 2016) was evaluated in two steps. First,
where, presumably, the scale is adapted to the work context [see we examined MI across the three Norwegian working samples.
for instance Olafsen et al. (2017), Silva et al. (2017)]. Moreover, Second, we examined the MI between the Norwegian and the
Gillet et al. (2012) have adapted the same scale to work and English scale. (3) The intercorrelations and reliability of the
validated it in French, and this has been used in several studies Norwegian and English subscales were then examined. (4) The
[see for instance Landry et al. (2016), Trépanier et al. (2016), criterion-related validity of the Norwegian and English scales
Trépanier et al. (2015)]. In addition, a few studies have applied was examined by looking at associations between the subscales
the Work-Related Basic Need Satisfaction Scale (Van den Broeck and important antecedents and outcomes of satisfaction and
et al., 2010) to assess basic psychological need frustration. In frustration of the basic psychological needs in the literature.
particular, in the studies by Vander Elst et al. (2012) and Van den In line with SDT and previous research, we examined the
Broeck et al. (2014) the reversed score items for need satisfaction associations between need satisfaction and frustration on
were used as indicators of need frustration. These measurements one side and managerial need support, autonomous work
of need frustration have some challenges. First, the scale motivation, controlled work motivation, vigor, emotional
developed by Bartholomew et al. (2011a) has been criticized for exhaustion, affective commitment, and turnover intention on the
also assessing antecedents of need satisfaction. Second, this scale other. In particular, we expected positive associations between
has not been stringently validated in the work domain [with the need satisfaction and managerial need support, autonomous
potential exception of Gillet et al. (2012), but this is not readily work motivation, vigor, and affective commitment, while
available to most researchers because it is in French]. Third, as negative associations between need satisfaction and controlled
past studies have proven need satisfaction and need frustration motivation, emotional exhaustion, and turnover intention. For
to be distinct concepts, assessing need frustration using the same need frustration, we expected the opposite pattern of associations
instrument as for need satisfaction is questionable. with the mentioned variables.

Frontiers in Psychology | www.frontiersin.org 3 July 2021 | Volume 12 | Article 697306

Olafsen et al. Basic Psychological Needs Work Scale

TABLE 1 | Overview of demographic characteristics of the study samples.

Sample 1 Sample 2 Sample 3 Sample 4

N 281 299 459 393

Gender
Men 43.1% 2.0% 21.0% 47.6%
Women 54.4% 96.0% 76.0% 52.4%
Other/unknown 2.5% 2.0% 3.0% 0%
Age Under 34 years: 27.6% Under 20 years: 0.0% Under 20 years: 0.7% Under 20 years: 0.3%
35–49 years: 44.0% 20–29 years: 20.5% 20–29 years: 9.3% 20–29 years: 17.0%
50 years and over: 28.4% 30–39 years: 23.9% 30–39 years: 21.2% 30–39 years: 39.7%
40–49 years: 23.9% 40–49 years: 32.1% 40–49 years: 24.4%
50–59 years: 18.2% 50–59 years: 26.9% 50–59 years: 12.7%
60 years and over: 13.5% 60 years and over: 9.9% 60 years and over: 5.9%

METHODS This translated version was then back-translated, and the original
scale was compared with this back-translation to ensure that
Procedures and Participants the items reflected their original content. After verifying that
Data were collected through online questionnaires. We used the meaning of the items was not changed in the translated
four samples to cover various occupations, organizations, and Norwegian version, the items in both Norwegian and English
the Norwegian and English language (total N = 1,432) in were adapted to fit the work context. The English version of
the validation of the BPNSFWS. Sample 1 consisted of 281 the scale appears in Appendix B. The instrument consisted of
employees in the finance and sales sector. Sample 2 consisted subscales for autonomy satisfaction (four items; e.g., I feel that the
of 299 dental hygienists. Sample 3 consisted of 459 employees decisions I make at work reflect what I really want), competence
in a municipality. For samples 1 to 3, participants received satisfaction (four items; e.g., I feel confident that I can do things
an invitation to the survey through their work e-mail. In the well at work), and relatedness satisfaction (four items; e.g., I
invitation they were informed about the purpose and content feel that the people I care about at work also care about me).
of the study, the estimated time-usage, that their participation The items were reported on a scale ranging from 1 (completely
was voluntary, and that their answers would be completely disagree) to 7 (completely agree).
confidential. Sample 4 consisted of 513 US employees reached
through Amazon Mechanical Turk. The respondents were
Need Frustration
informed that participation was voluntary and anonymous and
Need frustration was measured using an adapted version of
about the approximate response time. Responses were rewarded
the BPNSFS (Chen et al., 2015). The process was the same as
with $1,5 USD. In an effort to ensure high data quality, three
described for the need of satisfaction part of the scale. The
analyses were performed. First, two cases were removed as
instrument consisted of subscales autonomy frustration (four
duplicates from the same respondent. Second, 40 cases were
items; e.g., I feel pressured to do many of the things I do at
removed for being completed in <5 min, indicating insufficient
work), competence frustration (four items; e.g., I seriously doubt
time dedicated to the task. Finally, 78 cases were removed for
whether I can do things well at work), and relatedness frustration
having impossible inconsistencies in the replies, for example by
(four items; e.g., At work I feel excluded from the group that I
answering two directly opposing statements with “Completely
want to be a part of). The items were reported on a scale ranging
agree.” This left a sample of 393 respondents. Samples 1, 2, and 3
from 1 (completely disagree) to 7 (completely agree).
were used to validate the Norwegian Scale, while sample 4 were
used to validate the English scale. Demographics for the four
Managerial Need Support
samples are presented in Table 1. Approval was obtained from
Managerial need support was assessed in all four samples using
the Norwegian Center for Research Data prior to data collection
the six-item version of the Work Climate Questionnaire (Baard
(project numbers 578437, 53264, and 52866).
et al., 2004). The items (e.g., I feel understood by my manager; α
= 0.94, 0.94, 0.94, 0.94, ω = 0.94, 0.94, 0.95, 0.94 for samples 1, 2,
Measures 3, and 4, respectively), were measured on a scale ranging from 1
Need Satisfaction (completely disagree) to 7 (completely agree).
Need satisfaction was measured using an adapted version of the
BPNSFS (Chen et al., 2015) where the respondents are asked Motivation
to rate their level of agreement with the various scale items, The Norwegian and English version of the Multidimensional
in this case, as they related to their personal experiences at Work Motivation Scale (MWMS; Gagné et al., 2015) presented
work. The questionnaire adaptation process was guided by the participants with the following stem in samples 2, 3, and 4: “I
suggestions of the International Test Commission (Bartram et al., put effort into my job. . . ” Participants rated preselected responses
2018). First, the original scale was translated into Norwegian. that assessed external regulation (six items; e.g. Because others

Frontiers in Psychology | www.frontiersin.org 4 July 2021 | Volume 12 | Article 697306

Olafsen et al. Basic Psychological Needs Work Scale

will reward me financially only if I put enough effort into my MI was tested in four steps: (1) configural, (2) metric, (3)
job), introjected regulation (four items; e.g., Because I have to scalar, (4) residual (Putnick and Bornstein, 2016). further
prove to myself that I can), identified regulation (three items; e.g., evaluation of convergent and discriminant validity was made
Because putting effort into this job has personal significance to on the basis of calculation of the Average Variance Extracted
me), and intrinsic motivation (three items; e.g., Because what I do (AVE; Fornell and Larcker, 1981) and heterotrait-monotrait
in my work is exciting). Responses were made on a 7-point scale ratio (HTMT; Henseler et al., 2015) of correlations (i.e., the
from 1 (not at all for this reason) to 7 (exactly for this reason). average of the correlations of indicators across constructs
The scores on external regulation and introjection were added to measuring different phenomena relative to the average of the
make a composite for controlled work motivation (α = 0.85, 0.84, correlations of indicators within the same construct) among the
0.82, ω = 0.83, 0.82, 0.81 for samples 2, 3, and 4, respectively), six scale dimensions. Internal consistency was evaluated by score
while identified regulation and intrinsic motivation were added reliability using SPSS Statistics 25 and JASP. Finally, zero-order
to make a composite for autonomous work motivation (α = correlations were used to establish criterion validity (Kline, 2005)
0.90, 0.89, 0.92, ω = 0.89, 0.87, 0.92 for samples 2, 3, and between the subscales in the Norwegian and English BPNSFWS
4, respectively). with related variables.
While Appendix A shows that the items are mostly normally
Vigor distributed, a test of multivariate normality conducted by
In samples 2, 3, and 4, the vigor subscale of the short version calculating Mardia’s coefficient using the DeCarlo (1997) macro
of the Utrecht Work Engagement Scale (Schaufeli et al., 2006) showed evidence of multivariate non-normality. That is, Mardia’s
assessed vigor related to work (three items; e.g., At my work, I normalized (i.e., standardized) coefficients of kurtosis of 46.935
feel bursting with energy; α = 0.94, 0.89, 0.89, ω = 0.95, 0.89, in sample 1, 93.892 in sample 2, 89.147 in sample 3, and 100.473
0.89 for samples 2, 3, and 4, respectively). Responses were made in sample 4 were well-above the recommended cut-off of |3.0|
on a 7-point scale from 1 (never) to 7 (daily). suggested by Bentler and Wu (2002). Thus, the analyses were run
with the robust maximum likelihood estimator to account for the
Emotional Exhaustion
non-normal data, and the Satorra–Bentler (S–B) scaled χ2 and
In samples 2, 3, and 4, the emotional exhaustion subscale of
robust standard errors adjustment to the maximum likelihood
the Maslach Burnout Inventory (Maslach et al., 1996) assessed
estimator are reported in the results section.
emotional exhaustion at work (five items; e.g., I feel burned out
The fit of the models was evaluated using the chi-squared test
from my work; α = 0.90, 0.91, 0.96, ω = 0.90, 0.91, 0.95 for
(χ2 ), the root-mean-square error of approximation (RMSEA),
samples 2, 3, and 4, respectively). Responses were made on a
the comparative fit index (CFI), and the standardized root-mean-
7-point scale from 1 (never) to 7 (every day).
square residual (SRMR) as recommended by Hu and Bentler
Affective Commitment (1999). While values for CFI above 0.95 are recommended (Hu
In samples 2, and 4, affective occupational commitment was and Bentler, 1999), values above 0.90 were deemed acceptable
assessed using the Affective Commitment Scale developed by (Hoyle, 1995). Values below 0.08 were deemed acceptable for
Allen and Meyer (1990). The responses to the eight items (e.g., both SRMR and RMSEA (Hu and Bentler, 1999). MI was
This occupation has a great deal of personal meaning for me; α claimed acceptable if changes in the CFI were <0.01 coupled
= 0.84, 0.83, ω = 0.83, 0.89 for samples 2 and 4, respectively), with changes in RMSEA <0.015 and SRMR <0.030 (metric
were reported on a scale ranging from 1 (completely disagree) to invariance) or <0.015 (scalar and strict invariance) (Little, 2013).
7 (completely agree). If the constrained model was rejected, a less restrictive model
of partial invariance was evaluated in which, in accordance
Turnover Intention with modification indices and analysis of parameter estimates,
Turnover intentions over the past year were assessed using three equality constraints on one or more items were relaxed. If the
items based on the scale by Luchak and Gellatly (2007). A sample model of partial invariance was accepted using these criteria, it
item is “The past year, I have regularly had an intention to leave” was considered as the new reference model. For the remaining
(α = 0.88, 0.93, 0.91, 0.96, ω = 0.87, 0.93, 0.91, 0.96 for samples test of validity and reliability, values of Cronbach’s alpha (α)
1, 2, 3, and 4, respectively). In samples 2 and 4, three additional and MacDonald’s omega (ω) above 0.7 are typically deemed
items of current thinking about turnover from O’Driscoll and satisfactory for internal consistency (Nunnally, 1978), while
Beehr (1994) were assessed. A sample item is “I am thinking of AVE values of at least 0.50 (Fornell and Larcker, 1981) and
leaving this job” (α = 0.92, 0.97, ω = 0.93, 0.97 for samples 2 and HTMT ratio of correlation under 0.85 (Clark and Watson,
4, respectively). Responses were made on a 7-point scale from 1 1995; Klein, 2015) or 0.90 (Gold et al., 2001; Teo et al.,
(never) to 7 (always). 2008) are common guidelines for convergent and discriminant
validity, respectively.
Data Analyses
The factor structure of the translated and adapted BPNSFWS RESULTS
was examined with CFA in Mplus (Muthén and Muthén,
1998–2017). Multi-group CFA was performed in Mplus to Factor Structure
test for MI across the three Norwegian samples as well as For all samples, Table 2 goes to show that the six-factor model
between the Norwegian and the English version of the scale. fitted the data better than the alternative three-factor model with

Frontiers in Psychology | www.frontiersin.org 5 July 2021 | Volume 12 | Article 697306

Olafsen et al. Basic Psychological Needs Work Scale

TABLE 2 | Analyses of factor structure within the BPNSFWS across samples.

Model SBχ2 df p SCF RSMEA 90% CI CFI SRMR SBχ2 diff df diff

3-factor model Sample 1 944.44 243 <0.001 1.2523 0.104 0.097–0.111 0.76 0.120 – –
6-factor model Sample 1 425.57 231 <0.001 1.2380 0.056 0.048–0.064 0.93 0.059 429.35*** 12
3-factor model Sample 2 703.48 245 <0.001 1.5335 0.082 0.075–0.089 0.83 0.096 – –
6-factor model Sample 2 344.86 233 <0.001 1.5459 0.041 0.032–0.050 0.96 0.062 422.10*** 12
3-factor model Sample 3 968.28 245 <0.001 1.4412 0.083 0.078–0.189 0.80 0.086 – –
6-factor model Sample 3 462.23 233 <0.001 1.4446 0.048 0.042–0.055 0.94 0.053 529.20*** 12
3-factor model Sample 4 2,086.96 244 <0.001 1.5298 0.137 0.132–0.142 0.66 0.190 – –
6-factor model Sample 4 480.76 232 <0.001 1.5626 0.052 0.046–0.059 0.95 0.067 2725.79*** 12

***p < 0.001.

TABLE 3 | Invariance analysis of the BPNSFWS across the Norwegian samples.

Model SBχ2 df p SCF RMSEA 90% CI CFI SRMR SBχ2 diff df diff 1RMSEA 1CFI 1SRMR

Model A 1,225.34 697 <0.001 1.4100 0.048 0.044–0.053 0.943 0.057 – – – – –

Model B 1,314.35 733 <0.001 1.4189 0.049 0.045–0.054 0.937 0.065 86.22*** 36a 0.001 0.006 0.008
Model C 1,428.56 769 <0.001 1.4004 0.051 0.047–0.056 0.928 0.066 132.48*** 36b 0.002 0.009 0.001
Model D 1,624.41 817 <0.001 1.4574 0.055 0.051–0.059 0.912 0.067 154.75*** 48c 0.004 0.016 0.001
Model D2 1,565.51 815 <0.001 1.4547 0.053 0.049–0.057 0.919 0.066 117.16*** 46c 0.002 0.009 0.000

Model A: one-factor configural invariance (CI). Model B: one-factor CI and metric invariance (MI). Model C: one-factor CI, MI, and scalar invariance (SI). Model D: one-factor CI, MI, SI,
and invariant uniquenesses (IU). Model D2: one-factor CI, MI, SI, and partial IU.
***p <0.001.
a The reference model is Model A.
b The reference model is Model B.
c The reference model is Model C.

TABLE 4 | Invariance analysis of the BPNSFWS between the Norwegian and English scale.

Model SBχ2 df p SCF RMSEA 90% CI CFI SRMR SBχ2 diff df diff 1RMSEA 1CFI 1SRMR

Model A 1,052.65 461 <0.001 1.5380 0.043 0.040–0.047 0.957 0.051 – – – – –

Model B 1,145.45 479 <0.001 1.5418 0.045 0.042–0.049 0.951 0.059 89.73*** 18a 0.002 0.006 0.008
Model C 1,314.02 492 <0.001 1.5117 0.049 0.047–0.056 0.940 0.060 547.27*** 13b 0.004 0.011 0.001
Model C2 1,263.57 491 <0.001 1.5132 0.048 0.045–0.051 0.944 0.060 392.86*** 12b 0.003 0.007 0.001
Model D 1,433.03 520 <0.001 1.5621 0.051 0.048–0.054 0.933 0.062 136.61*** 29c 0.003 0.011 0.002
Model D2 1,398.05 519 <0.001 1.5623 0.050 0.047–0.053 0.936 0.062 112.30*** 28c 0.002 0.008 0.002

Model A: one-factor configural invariance (CI). Model B: one-factor CI and metric invariance (MI). Model C: one-factor CI, MI, and scalar invariance (SI). Model C2: one-factor CI, MI,
and partial SI. Model D: one-factor CI, MI, partial SI, and invariant uniquenesses (IU). Model D2: one-factor CI, MI, SI, and partial IU.
***p <0.001.
a The reference model is Model A.
b The reference model is Model B.
c The reference model is Model C.

acceptable fit indices for Sample 11 : χ2 (df = 232) = 433.36, p CI (0.049, 0.065); Sample 22 : χ2 (df = 233) = 344.86, p < 0.001,
< 0.001, CFI = 0.93, SRMR = 0.060, and RMSEA = 0.057, 90% CFI = 0.96, SRMR = 0.062, and RMSEA = 0.041, 90% CI (0.032,

1 Modification indices suggested adding residual covariances between some of the

indicators for the six factors. In sample 1, 6 residual covariances were added 2 Modification indices suggested adding residual covariances between some of the

(competence satisfaction item 1 with competence satisfaction item 2, relatedness indicators for the six factors. In sample 2, 4 residual covariances were added
satisfaction item 1 with relatedness satisfaction item 3, autonomy satisfaction item (competence satisfaction item 1 with competence satisfaction item 4, autonomy
1 with autonomy satisfaction item 3, autonomy frustration item 2 with autonomy frustration item 1 with autonomy frustration item 4, relatedness frustration
frustration item 3, competence frustration item 2 with competence frustration item 1 with relatedness frustration item 2, autonomy satisfaction item 2 with
item 4, relatedness frustration item 1 with relatedness frustration item 2). These autonomy satisfaction item 3). These residual covariances were included in the
residual covariances were included in the test of MI and represented group-specific test of MI and represented group-specific parameters, meaning that partial MI
parameters, meaning that partial MI is tested. is tested.

Frontiers in Psychology | www.frontiersin.org 6 July 2021 | Volume 12 | Article 697306

Olafsen et al. Basic Psychological Needs Work Scale

0.050); Sample 33 : χ2 (df = 230) = 414.66, p < 0.001, CFI = 0.95, did cause a significant reduction in fit compared with Model B:
SRMR = 0.052, and RMSEA = 0.043, 90% CI (0.037, 0.050); and S–B χ2diff = 132.48 (1df = 36), p < 0.001, it was considered
Sample 44 : χ2 (df = 232) = 480.76, p < 0.001, CFI = 0.95, SRMR acceptable because 1CFI was <0.01, and 1RMSEA and 1SRMR
= 0.067, and RMSEA = 0.052, 90% CI (0.046, 0.059), respectively. were <0.015.
Furthermore, all items had significant loadings (ranging from
0.60 to 0.90, p <0.001, with an average loading of 0.77 in Sample Invariant Uniqueness Test
1; 0.51 to 0.94, p < 0.001, with an average loading of 0.78 in A model adding cross-group equality constraints on all like items’
Sample 2; 0.58 to 0.86, p < 0.001, with an average loading of residual variance was analyzed (Model D). The results provided
0.74 in Sample 3; and 0.72 to 0.92, p < 0.001, with an average in Table 3 indicated acceptable fit for the tested model: χ2 (df =
loading of 0.84 in Sample 4) on their intended latent factor (see 817) = 1,624.41, p < 0.001, CFI = 0.912, SRMR = 0.067, and
Appendix B). RMSEA = 0.055, 90% CI (0.051, 0.059). Compared with Model
C, the constraints did cause a significant reduction in fit: S–B
Measurement Invariance Among the χ2diff = 154.75 (1df = 48), p < 0.001 and 1CFI was >0.01,
Norwegian Samples while 1RMSEA and 1SRMR were <0.015. The modification
Configural Invariance Test indices suggested freely estimating the residual of item 2 for
To examine the configural MI of the measurement scale, a competence frustration. The new partial invariant uniqueness
simultaneous multi-group CFA of the six-factor model was tested model (Model D2) showed acceptable fit to the data: χ2 (df
in the three Norwegian samples. This model (Model A) imposed = 815) = 1,565.51, p < 0.001, CFI = 0.919, SRMR = 0.066,
no equality constraints on parameter estimates across groups. and RMSEA = 0.053, 90% CI (0.049, 0.057). Even though the
The results provided in Table 3 indicated acceptable fit for the constraints did cause a significant reduction in fit compared with
tested model: χ2 (df = 697) = 1,225.34, p < 0.001, CFI = 0.943, Model C: S–B χ2diff = 117.16 (1df = 46), p < 0.001, it was
SRMR = 0.057, and RMSEA = 0.048, 90% CI (0.044, 0.053). considered acceptable because 1CFI was <0.01, and 1RMSEA
and 1SRMR were <0.015.
Metric Invariance Test
The same model was tested simultaneously in the three Measurement Invariance Between the
Norwegian samples but constraining the corresponding item Norwegian and the English Scale
slopes to be equal across groups (Model B). The results provided Before testing the four-steps of measurement invariance of
in Table 3 indicated acceptable fit for the tested model: χ2 (df the six-factor model between the Norwegian and the English
= 733) = 1,314.35, p < 0.001, CFI = 0.937, SRMR = 0.065, samples, the three Norwegian samples were combined into a
and RMSEA = 0.049, 90% CI (0.045, 0.054). Even though the single sample. Results of a CFA indicated acceptable fit for the
constraints did cause a significant reduction in fit compared with tested model: χ2 (df = 231) = 637.10, p < 0.001, CFI = 0.951,
Model A: S–B χ2diff = 82.22 (1df = 36), p < 0.001, it was SRMR = 0.043, and RMSEA = 0.043, 90% CI (0.039, 0.046)5 .
considered acceptable because 1CFI was <0.01, 1RMSEA was
<0.015, and 1SRMR was <0.030. Configural Invariance Test
To examine the configural MI of the measurement scale, a
Scalar Invariance Test simultaneous multi-group CFA model was tested in between the
The same model was tested simultaneously in the three combined Norwegian sample (group 1) and the English sample
Norwegian samples but constraining both the corresponding (group 2). This model (Model A) imposed no equality constraints
item slopes and all the intercepts of the observed items to be on parameter estimates across groups. The results provided in
equal across groups (Model C). The results provided in Table 3 Table 4 indicated acceptable fit for the tested model: χ2 (df =
indicated acceptable fit for the tested model: χ2 (df = 769) = 461) = 1,052.65, p < 0.001, CFI = 0.957, SRMR = 0.051, and
1,428.56, p < 0.001, CFI = 0.928, SRMR = 0.066, and RMSEA RMSEA = 0.043, 90% CI (0.040, 0.047).
= 0.051, 90% CI (0.047, 0.056). Even though the constraints
Metric Invariance Test
3 Modification indices suggested adding residual covariances between some of the The same model was tested simultaneously in the two groups but
indicators for the six factors. In sample 3, 4 residual covariances were added constraining the corresponding item slopes to be equal across
(autonomy frustration item 2 with autonomy frustration item 3, relatedness
groups (Model B). The results provided in Table 4 indicated
satisfaction item 1 with relatedness satisfaction item 3, relatedness frustration
item 1 with relatedness frustration item 2, relatedness satisfaction item 2 with acceptable fit for the tested model: χ2 (df = 479) = 1,145.45, p <
relatedness satisfaction item 3). These residual covariances were included in the
test of MI and represented group-specific parameters, meaning that partial MI is 5 Modification indices suggested adding residual covariances between some of the

tested. indicators for the six factors. In the total Norwegian sample, 7 residual covariances
4 Modification indices suggested adding residual covariances between some of the were added (autonomy frustration item 2 with autonomy frustration item 3,
indicators for the six factors. In sample 4, 5 residual covariances were added relatedness satisfaction item 1 with relatedness satisfaction item 2, relatedness
(autonomy frustration item 1 with autonomy frustration item 4, relatedness frustration item 1 with relatedness frustration item 3, competence satisfaction
satisfaction item 1 with relatedness satisfaction item 3, relatedness frustration item 3 with competence satisfaction item 3, competence frustration item 1 with
item 1 with relatedness frustration item 4, competence frustration item 1 with competence frustration item 4, relatedness satisfaction item 2 with relatedness
competence frustration item 4, competence frustration item 3 with competence satisfaction item 3, relatedness satisfaction item 1 with relatedness satisfaction item
frustration item 4). These residual covariances were included in the test of MI and 2). These residual covariances were included in the test of MI and represented
represented group-specific parameters, meaning that partial MI is tested. group-specific parameters, meaning that partial MI is tested.

Frontiers in Psychology | www.frontiersin.org 7 July 2021 | Volume 12 | Article 697306

Olafsen et al. Basic Psychological Needs Work Scale

0.001, CFI = 0.951, SRMR = 0.059, and RMSEA = 0.045, 90% CI

−0.24

−0.55

−0.45

0.61

0.84

0.72
6. Relatedness frustration
(0.042, 0.049). Even though the constraints did cause a significant

S4

S, Sample. AVE reported on, HTMT ratio of correlations reported above, and intercorrelations among scale dimensions reported below the diagonal. ***p < 0.001, **p < 0.01, * p < 0.05. The bold values on the diagonal are AVE.
reduction in fit compared with Model A: S–B χ2diff = 89.73 (1df

−0.45 −0.55 −0.47

−0.09 −0.24 −0.14

−0.52 −0.77 −0.65

= 18), p < 0.001, it was considered acceptable because 1CFI was

0.55

0.48

0.65
S3
<0.01, 1RMSEA was <0.015, and 1SRMR was <0.030.

0.58

0.49

0.68
S2
Scalar Invariance Test

0.32

0.28

0.59
S1
The same model was tested simultaneously in the two groups
samples but constraining both the corresponding item slopes and

−0.54*** −0.52*** −0.80*** −0.69*** −0.42*** 0.57*** 0.57*** 0.57*** 0.62*** 0.53*** 0.49*** 0.46*** 0.84***
−0.10

−0.63

−0.20

0.44

0.73
5. Competence frustration

S4
all the intercepts of the observed items to be equal across groups
(Model C). The results provided in Table 4 indicated acceptable

−0.40

−0.68

−0.28

0.49

0.54
fit for the tested model: χ2 (df = 492) = 1,314.02, p < 0.001, CFI

S3
= 0.940, SRMR = 0.060, and RMSEA = 0.049, 90% CI (0.047,

−0.44

−0.69

−0.46

0.42

0.68
S2
0.056). The constraints did cause a significant reduction in fit
compared with Model B: S–B χ2diff = 547.27 (1df = 13), p <

−0.37

−0.58

−0.14

0.25

0.47
S1
0.001 and 1CFI was >0.01, while 1RMSEA and 1SRMR were
<0.015. The modification indices suggested freely estimating the

−0.42*** −0.30*** −0.19*** 0.50*** 0.30*** 0.53*** 0.47***

−0.62

−0.38

−0.52

0.66
S4
intercept of item 3 for relatedness satisfaction. The new partial

4. Autonomy frustration
invariant uniqueness model (Model C2) showed acceptable fit

−0.55

−0.14

−0.32

0.49
S3
to the data: χ2 (df = 419) = 1,263.57, p < 0.001, CFI = 0.944,
SRMR = 0.060, and RMSEA = 0.048, 90% CI (0.045, 0.051).

−0.64

−0.33

−0.47

0.58
S2
Even though the constraints did cause a significant reduction in
fit compared with Model B: S–B χ2diff = 392.86 (1df = 12), p <

−0.59

−0.15

−0.25

0.60
S1
0.001, it was considered acceptable because 1CFI was <0.01, and
1RMSEA and 1SRMR were <0.015.

−0.41*** −0.22** −0.44*** −0.35*** −0.53***

0.74

0.46

0.81
S4
Invariant Uniqueness Test 3. Relatedness satisfaction
A model adding cross-group equality constraints on all like items’

0.54

0.28

0.56
S3
residual variance was analyzed (Model D). The results provided
TABLE 5 | AVE, HRMT ratio of correlations, and intercorrelations among scale dimensions across samples.

in Table 4 indicated acceptable fit for the tested model: χ2 (df = 0.57

0.37

0.64
S2

520) = 1,433.03, p < 0.001, CFI = 0.933, SRMR = 0.062, and

RMSEA = 0.051, 90% CI (0.048, 0.054). Compared with Model

−0.15*
0.50

0.27

0.60
S1

C1, the constraints did cause a significant reduction in fit: S–

B χ2diff = 136.61 (1df = 29), p < 0.001, and 1CFI was >0.01,

−0.60*** −0.64*** −0.69*** −0.64***

0.44***

while 1RMSEA and 1SRMR were <0.015. The modification

0.41

0.64
S4
2. Competence satisfaction

indices suggested freely estimating the residual of item 4 for

relatedness frustration. The new partial invariant uniqueness
−0.13*

−0.14*
0.29***
0.43

0.55
S3

model (Model D2) showed acceptable fit to the data: χ2 (df

= 519) = 1,398.05, p < 0.001, CFI = 0.936, SRMR = 0.062,
−0.21***

−0.23***
0.36***
0.49

0.61

and RMSEA = 0.050, 90% CI (0.047, 0.053). Even though the

S2

constraints did cause a significant reduction in fit compared with

Model B: S–B χ2diff = 112.30 (1df = 28), p < 0.001, it was
−0.14*
0.26***

−0.09
0.45

0.67
S1

considered acceptable because 1CFI was <0.01, and 1RMSEA

and 1SRMR were <0.015.
−0.59*** −0.62*** −0.51*** −0.61***

−0.45*** −0.56*** −0.41*** −0.20***

0.38***

0.72***

−0.07
0.72
S4
1. Autonomy satisfaction

Intercorrelations and Reliability

−0.38*** −0.41*** −0.38***
0.41***

0.52***
0.55

As displayed in Table 5, across the four samples, the latent

S3

variables of autonomy and competence satisfaction correlated on

0.46***

0.55***

average 0.43, autonomy and relatedness satisfaction correlated

0.54
S2

on average 0.56, and competence and relatedness satisfaction

correlated 0.34. Similarly, autonomy and competence frustration
0.46***

0.44***
0.64
S1

correlated on average 0.45, autonomy and relatedness frustration

correlated on average 0.58, and competence and relatedness
frustration correlated 0.58. As for the correlations between the
2. Competence

5. Competence
3. Relatedness

6. Relatedness
1. Autonomy

4. Autonomy

subscales of autonomy, competence, and relatedness, autonomy

satisfaction

satisfaction

satisfaction

frustration

frustration

frustration

satisfaction and autonomy frustration correlated on average

−0.58, competence satisfaction and competence frustration

Frontiers in Psychology | www.frontiersin.org 8 July 2021 | Volume 12 | Article 697306

Olafsen et al. Basic Psychological Needs Work Scale

TABLE 6 | Zero-order correlations among need satisfaction, need frustration, and work-related correlates across samples.

Autonomy satisfaction Competence satisfaction Relatedness satisfaction

S1 S2 S3 S4 S1 S2 S3 S4 S1 S2 S3 S4

Managerial need support 0.60*** 0.61*** 0.52*** 0.71*** 0.26*** 0.21*** 0.22*** 0.44*** 0.39*** 0.36*** 0.35*** 0.69***
Autonomous motivation 0.44*** 0.41*** 0.70*** 0.45*** 0.24*** 0.35*** 0.37*** 0.33*** 0.63***
Controlled motivation −0.09 0.06 0.29*** −0.04 −0.05 0.04 0.04 0.03 0.29***
Emotional exhaustion −0.37*** −0.31*** −0.41*** −0.25*** −0.12* −0.36*** −0.29** −0.14** −0.39***
Vigor 0.48*** 0.47*** 0.62*** 0.32*** 0.32*** 0.39*** 0.29*** 0.32*** 0.62***
Affective commitment 0.31*** 0.66*** 0.21*** 0.34*** 0.23*** 0.66***
Turnover intentions −0.46*** −0.42*** −0.38*** −0.50*** −0.14* −0.28*** −0.09 −0.31*** −0.21*** −0.38*** −0.29*** −0.45***

Autonomy frustration Competence frustration Relatedness frustration

Managerial need support −0.41*** −39*** −0.39*** −0.48*** −0.23*** −22*** −0.22*** −0.19*** −0.40*** −0.39*** −0.40*** −0.35***
Autonomous motivation −0.36*** −0.23*** −0.44*** −0.41 −0.20*** −0.07 −0.23*** −0.24*** −0.20***
Controlled motivation 0.16** 0.09 0.05 0.17** 0.21*** 0.28*** 0.10 0.18*** 0.17**
Emotional exhaustion 0.51*** 0.40*** 0.59*** 0.26*** 0.29*** 0.48*** 0.33*** 0.28*** 0.49***
Vigor −0.43*** −0.33*** −0.48*** −0.30*** −0.32*** −0.15** −0.22*** −0.31*** −0.26***
Affective commitment −0.24*** −0.54*** −0.19** −0.24*** −0.17** −0.39***
Turnover intentions 0.39*** 0.51*** 0.39*** 0.57*** 0.31*** 0.26*** 0.19*** 0.40*** 0.36*** 0.39*** 0.31*** 0.49***

S, Sample. ***p < 0.001, **p < 0.01, *p < 0.05.

correlated on average −0.64, and relatedness satisfaction and associated with controlled work motivation. Rather, need
relatedness frustration correlated on average −0.61. frustration was a better predictor of controlled work motivation
To further evaluate convergent and discriminant validity, despite some inconsistencies between the samples. There was
AVE and HTMT ratio was calculated. Results showed that AVE one exception of this, as the English sample showed a positive
for each of the six need dimensions had an AVE of 0.5 or correlation between autonomy need satisfaction and controlled
above (Fornell and Larcker, 1981) across the fours samples. Two work motivation, and between relatedness need satisfaction
exceptions were found; for autonomy frustration in sample 3 and controlled work motivation. Additional analysis revealed
(AVE = 0.49) and for competence frustration in sample 1 (AVE that this association was caused by a positive correlation to
= 0.47). However, these represent quite small deviation from introjection, in particular the introjection approach items
the recommendation and only appear in one sample and was (Gagné et al., 2015).
deemed acceptable. As for the HTMT ratio, none were above the Furthermore, as expected, frustration of the basic
strictest threshold of 0.85 (Clark and Watson, 1995; Klein, 2015) psychological needs for autonomy, competence, and relatedness
indicating discriminant validity for the six scale dimensions. was negatively associated with managerial need support,
Cronbach’s alpha (α) and McDonald’s ω were calculated for autonomous work motivation, vigor, and affective commitment,
each subscale and showed adequate reliability with an average α while positively associated with emotional exhaustion and
= 0.86; ω = 0.86 for autonomy satisfaction, α = 0.86; ω = 0.86 turnover intention.
for competence satisfaction, α = 0.87; ω = 0.87 for relatedness
satisfaction, α = 0.86; ω = 0.86 for autonomy frustration, α
= 0.86; ω = 0.85 for competence frustration, and α = 0.89; ω DISCUSSION
= 0.89 for relatedness frustration across the four samples (see
Appendix B). The purpose of the present study was to validate the BPNSFWS,
which is based on the theoretical framework of SDT (Deci
and Ryan, 2000; Ryan and Deci, 2017). The results across
Criterion-Related Validity three samples of a total of 1,432 employees provided acceptable
The correlations between the six sub-dimensions of the support for the properties of the adapted scale. In particular,
BPNSFWS and the criterion-related variables are presented in consistent with SDT, results showed support for a six-factor
Table 6. As expected, autonomy, competence, and relatedness structure of the BPNSFWS where satisfaction and frustration of
need satisfaction were positively associated with managerial the basic psychological needs for autonomy, competence, and
need support, autonomous work motivation, vigor, and affective relatedness were distinct factors (Chen et al., 2015). This factorial
commitment, while negatively associated with emotional structure held across three different Norwegian occupational
exhaustion and turnover intention. However, need satisfaction samples and one English sample. Calculations of AVE and
of autonomy, competence, and relatedness were not significantly HTMT ratios of correlations further supported the validity of

Frontiers in Psychology | www.frontiersin.org 9 July 2021 | Volume 12 | Article 697306

Olafsen et al. Basic Psychological Needs Work Scale

the measurement scale. Furthermore, the subscales for each et al., 2003). Future studies might examine whether other
need proved to have satisfactory internal consistency. With these methodological artifacts or personality factors may influence
results, the present study contributes to establishing an important responses to the scale. Second, the present study supports
tool for a central concept within SDT for future studies of the criterion-related validity of the BPNSFWS by means
organizational psychology based on this theoretical framework. of cross-sectional associations. Future studies may further
An important feature of the BPNSFWS is that it was adapted examine temporal stability of the measurement scale and causal
and validated specifically for the work context. Moreover, it relations between work-related need satisfaction and need
addresses need satisfaction and need frustration as separate frustration and its antecedents and consequences by means
dimensions. Because previous research has employed domain- of longitudinal or experimental studies. In addition, diary
general scales to measure the basic psychological needs and/or studies can be used to focus on intra-individual differences in
scales with notable limitations such as content validity and lack need satisfaction and frustration and their correlates. Third,
of rigorous validation, the adapted scale enables researchers to the present study included convenience samples of different
study need satisfaction and frustration at work with items that organization- or occupation-specific samples. Future research
are relevant in the work context and meaningful to employees in different sectors and countries may further add to the
and, therefore, have a better chance of generating reliable generalizability of the findings. Moreover, while the present
and valid scores of employee need satisfaction and frustration study, which was adapted and validated in Norwegian and
at work. English represents the first step toward a valid measurement
The associations of need satisfaction and frustration with instrument for use in the study of the work-related question
common antecedents and outcomes within the SDT literature in organizational psychology within the framework of SDT,
in the work domain generally provided criterion-related validity validation of the scale in other languages will be fruitful.
of the adapted scale. In particular, satisfaction of the three Lastly, in recent years the concept of need unfulfillment
basic psychological needs was found to be positively related has emerged in the literature accounting for a possible
to positive work correlates (i.e., managerial need support, third dimension of need states of relevance in considering
autonomous work motivation, vigor, and affective commitment) the role of basic psychological needs in human motivation,
and negatively related to negative work correlates (i.e., emotional functioning, and wellness (Huyghebaert-Zouaghi et al., 2020).
exhaustion and turnover intention). Conversely, frustration Adding this dimension to the BPNSFWS to establish a
of the basic psychological needs was found to be negatively common measurement of also this possible dimension of
related to positive work correlates (i.e., managerial need the basic psychological need can be worthwhile in future
support, autonomous work motivation, vigor, and affective research efforts.
commitment) and positively related to negative work correlates
(i.e., controlled work motivation, emotional exhaustion, and Conclusion
turnover intention). In the English sample, albeit less strongly SDT offers a framework for understanding motivational
than to autonomous work motivation, autonomy and relatedness processes at work where fulfillment or frustration of employees’
need satisfaction had a positive correlation to controlled work basic psychological needs are highlighted as key facilitators
motivation. As this was caused by a positive correction to of employees’ motivation, work functioning, and well-
introjection, it follows previous findings for associations to the being. The current study provides researchers with a tool
regulations in the MWMS where introjection also previously to assess employees’ basic psychological need satisfaction and
has been found to relate to “positive” variables (Gagné et al., frustration at work by adapting the current go-to measurement
2015). instrument for basic psychological need satisfaction and need
Because most previous research has focused on the bright frustration to the work context. In addition, it validates the
path of motivational processes at work (Deci et al., 2017), the BPNSFWS in Norwegian and English. Relying on a common
validation of the BPNSFWS is of particular importance for future operationalization of need satisfaction and frustration in
studies on the dark side of these processes by providing a valid future research enables comparisons across studies, cultures,
instrument to assess this path. As seen from the correlations in and contexts that contribute to a more unified development
Table 2, need frustration is important in the association with of this field. Therefore, the current study contributes to a
controlled forms of motivation that have received less attention in unifying measurement of one of the central underpinnings
the literature, maybe because it has been challenging to predict by of SDT.
need satisfaction. Need frustration is, therefore, important to take
into consideration when assessing the dark side of motivational DATA AVAILABILITY STATEMENT
processes at work.
The raw data supporting the conclusions of this article will be
Limitations and Future Research Directions made available by the authors upon reasonable request.
Some potential limitations may exist in the present study.
First, because the data were collected relying only on the ETHICS STATEMENT
BPNSFWS self-report measure to assess the internal process
of need satisfaction and frustration, the results may have The studies involving human participants were reviewed
been influenced by common method variance (Podsakoff and approved by Norwegian Center for Research Data. The

Frontiers in Psychology | www.frontiersin.org 10 July 2021 | Volume 12 | Article 697306

Olafsen et al. Basic Psychological Needs Work Scale

patients/participants provided their written informed consent to writing up the manuscript. CWF: contributed in data collection
participate in this study. and to writing up the manuscript.

SUPPLEMENTARY MATERIAL
AUTHOR CONTRIBUTIONS
The Supplementary Material for this article can be found
AO: responsible for collection of data, data analysis, and writing online at: https://www.frontiersin.org/articles/10.3389/fpsyg.
up the manuscript. HH: contributed in data collection and to 2021.697306/full#supplementary-material

REFERENCES Gagné, M., Forest, J., Vansteenkiste, M., Crevier-Braud, L., Van den Broeck, A.,
Aspeli, A., et al. (2015). The multidimensional work motivation scale: validation
Allen, N. J., and Meyer, J. P. (1990). The measurement and antecedents of affective, evidence in seven languages and nine countries. Euro. J. Work Organ. Psychol.
continuance and normative commitment to the organization. J. Occup. Psychol. 24, 178–196. doi: 10.1080/1359432X.2013.877892
63, 1–18. doi: 10.1111/j.2044-8325.1990.tb00506.x Gillet, N., Fouquereau, E., Lequeurre, J., Bigot, L., and Mokounkolo,
Baard, P. P., Deci, E. L., and Ryan, R. M. (2004). Intrinsic need satisfaction: a R. (2012). Validation d’une échelle de frustration des besoins
motivational basis of performance and well-being in two work settings. J. Appl. psychologiques au travail (EFBPT). Psychol. Trav. Organ. 18, 328–344.
Soc. Psychol. 34, 2045–2068. doi: 10.1111/j.1559-1816.2004.tb02690.x doi: 10.1016/S1420-2530(16)30074-7
Bartholomew, K. J., Ntoumanis, N., Ryan, R. M., Bosch, J. A., and Thøgersen- Gold, A. H., Malhotra, A., and Segars, A. H. (2001). Knowledge management: an
Ntoumani, C. (2011b). Self-Determination Theory and diminished organizational capabilities perspective. J. Manage. Informat. Syst. 18, 185–214.
functioning: the role of interpersonal control and psychological need thwarting. doi: 10.1080/07421222.2001.11045669
Pers. Soc. Psychol. Bull. 37, 1459–1473. doi: 10.1177/0146167211413125 Henseler, J., Ringle, C. M., and Sarstedt, M. (2015). A new criterion for assessing
Bartholomew, K. J., Ntoumanis, N., Ryan, R. M., and Thøgersen-Ntoumani, discriminant validity in variance-based structural equation modeling. J. Acad.
C. (2011a). Psychological need thwarting in the sport context: assessing Market. Sci. 43, 115–135. doi: 10.1007/s11747-014-0403-8
the darker side of athletic experience. J. Sport Exerc. Psychol. 33, 75–102. Hoyle, R. H. (1995). Structural Equation Modeling. Thousand Oaks, CA: Sage.
doi: 10.1123/jsep.33.1.75 Hu, L., and Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance
Bartram, D., Berberoglu, G., Grégoire, J., Hambleton, R., Muniz, J., and van de structure analysis: conventional criteria versus new alternatives. Struct. Equat.
Vijver, F. (2018). ITC guidelines for translating and adapting tests (second Model. Multidiscipl. J. 6, 1–55. doi: 10.1080/10705519909540118
edition). Int. J. Test. 18, 101–134. doi: 10.1080/15305058.2017.1398166 Huyghebaert-Zouaghi, T., Ntoumanis, N., Berjot, S., and Gillet, N. (2020).
Baumeister, R. F., and Leary, M. R. (1995). The need to belong: desire for Advancing the conceptualization and measurement of psychological need
interpersonal attachments as a fundamental human motivation. Psychol. Bull. states: a 3 × 3 model based on self-determination theory. J. Care. Assess.
117, 497–529. doi: 10.1037/0033-2909.117.3.497 1069072720978792. doi: 10.1177/1069072720978792
Bentler, P. M., and Wu, E. J. (2002). EQS for Windows User’s Guide. Encino, CA: Klein, R. B. (2015). Principles and Practice of Structural Equation Modeling. New
Multivariate Software. York, NY: Guilford Publications
Chen, B., Vansteenkiste, M., Beyers, W., Boone, L., Deci, E. L., Van der Kline, T. J. (2005). Psychological Testing: A Practical Approach to Design and
Kaap-Deeder, J., et al. (2015). Basic psychological need satisfaction, need Evaluation. Thousand Oaks, CA: Sage. doi: 10.4135/9781483385693
frustration, and need strength across four cultures. Motiv. Emot. 39, 216–236. Landry, A. T., Kindlein, J., Trépanier, S. G., Forest, J., Zigarmi, D., Houson,
doi: 10.1007/s11031-014-9450-1 D., et al. (2016). Why individuals want money is what matters: using self-
Clark, L. A., and Watson, D. (1995). Constructing validity: basic issues in objective determination theory to explain the differential relationship between motives
scale development. Psychol. Assess. 7, 309–319. doi: 10.1037/1040-3590.7.3.309 for making money and employee psychological health. Motiv. Emot. 40,
Cordeiro, P., Paixao, P., Lens, W., Lacante, M., and Sheldon, K. (2016). Factor 226–242. doi: 10.1007/s11031-015-9532-8
structure and dimensionality of the balanced measure of psychological needs Little, T. D. (2013). Longitudinal Structural Equation Modeling. New York, NY: The
among Portuguese high school students: relations to well-being and ill-being. Guilford Press.
Learn. Individ. Differ. 47, 51–60. doi: 10.1016/j.lindif.2015.12.010 Longo, Y., Gunz, A., Curtis, G. J., and Farsides, T. (2016). Measuring need
DeCarlo, L. T. (1997). On the meaning and use of kurtosis. Psychol. Methods 2, satisfaction and frustration in educational and work contexts: The Need
292–307. doi: 10.1037/1082-989X.2.3.292 Satisfaction and Frustration Scale (NSFS). J. Happiness Stud. 17, 295–317.
deCharms, R. (1968). Personal Causation. New York, NY: Academic Press. Luchak, A. A., and Gellatly, I. R. (2007). A comparison of linear and nonlinear
Deci, E. L., Olafsen, A. H., and Ryan, R. M. (2017). Self-determination theory in relations between organizational commitment and work outcomes. J. Appl.
work organizations: state of the science. Annu. Rev. Organ. Psychol. Organ. Psychol. 92, 786–793. doi: 10.1037/0021-9010.92.3.786
Behav. 4, 19–43. doi: 10.1146/annurev-orgpsych-032516-113108 Martinent, G., Guillet-Descas, E., and Moiret, S. (2015). Reliability and
Deci, E. L., and Ryan, R. M. (2000). The “what” and “why” of goal pursuits: validity evidence for the french psychological need thwarting scale
human needs and the self-determination of behavior. Psychol. Inq. 11, 227–268. (PNTS) scores: significance of a distinction between thwarting and
doi: 10.1207/S15327965PLI1104_01 satisfaction of basic psychological needs. Psychol. Sport Exerc. 20, 29–39.
Deci, E. L., Ryan, R. M., Gagné, M., Leone, D. R., Usunov, J., and Kornazheva, B. P. doi: 10.1016/j.psychsport.2015.04.005
(2001). Need satisfaction, motivation, and well-being in the work organizations Maslach, C., Jackson, S. E., and Leiter, M. P. (1996). Maslach Burnout Inventory
of a former eastern bloc country: a cross-cultural study of self-determination. Manual. Palo Alto, CA: Consulting Psychologists Press.
Pers. Soc. Psychol. Bull. 27, 930–942. doi: 10.1177/0146167201278002 Muthén, L. K., and Muthén, B. O. (1998–2017). Mplus User’s Guide, Vol. 8. Los
Fornell, C., and Larcker, D. F. (1981). Evaluating structural equation models Angeles, CA: Muthén & Muthén.
with unobserved variables with measurement error. J. Market. Res. 18, 39–50. Nunnally, J. C. (1978). Psychometric Theory. New York, NY: McGraw-Hill
doi: 10.1177/002224378101800104 O’Driscoll, M. P., and Beehr, T. A. (1994). Supervisor behaviors, role stressors
Gagné, M., and Deci, E. L. (2014). “The history of self-determination theory in and uncertainty as predictors of personal outcomes for subordinates. J. Organ.
psychology and management,” in The Oxford Handbook of Work Engagement, Behav. 15, 141–155. doi: 10.1002/job.4030150204
Motivation and Self-Determination Theory, ed M. Gagné (New York, Olafsen, A. H., and Deci, E. L. (2020). Self-determination theory
NY: Oxford University Press), 1–9. doi: 10.1093/oxfordhb/9780199794911. and its relation to organizations. Oxford Encycl. Psychol.
013.006 doi: 10.1093/acrefore/9780190236557.013.112

Frontiers in Psychology | www.frontiersin.org 11 July 2021 | Volume 12 | Article 697306

Olafsen et al. Basic Psychological Needs Work Scale

Olafsen, A. H., and Halvari, H. (2017). Motivational mechanisms in the relation satisfaction and frustration. Euro. J. Work Organ. Psychol. 25, 690–706.
between job characteristics and employee functioning. Span. J. Psychol. 20:E38. doi: 10.1080/1359432X.2015.1132200
doi: 10.1017/sjp.2017.34 Trépanier, S. G., Forest, J., Fernet, C., and Austin, S. (2015). On the
Olafsen, A. H., Halvari, H., Forest, J., and Deci, E. L. (2015). Show them the money? psychological and motivational processes linking job characteristics to
The role of pay, managerial need support, and justice in a self-determination employee functioning: insights from self-determination theory. Work Stress 29,
theory model of intrinsic work motivation. Scand. J. Psychol. 56, 447–457. 286–305. doi: 10.1080/02678373.2015.1074957
doi: 10.1111/sjop.12211 Van den Broeck, A., Ferris, D. L., Chang, C.-H., and Rosen, C. C. (2016). A review
Olafsen, A. H., Niemiec, C. P., Halvari, H., Deci, E. L., and Williams, of self-determination theory’s basic psychological needs at work. J. Manage.
G. C. (2017). On the dark side of work: a longitudinal analysis using Account. Res. 42, 1195–1229. doi: 10.1177/0149206316632058
self-determination theory. Euro. J. Work Organ. Psychol. 26, 275–285. Van den Broeck, A., Sulea, C., Vander Elst, T., Fischmann, G., Iliescu, D., and
doi: 10.1080/1359432X.2016.1257611 De Witte, H. (2014). The mediating role of psychological needs in the relation
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., and Podsakoff, N. P. (2003). between qualitative job insecurity and counterproductive work behavior. Care.
Common method biases in behavioral research: a critical review of the Dev. Int. 19, 526–547. doi: 10.1108/CDI-05-2013-0063
literature and recommended remedies. J. Appl. Psychol. 88, 879–903. Van den Broeck, A., Vansteenkiste, M., De Witte, H., Soenens, B., and Lens,
doi: 10.1037/0021-9010.88.5.879 W. (2010). Capturing autonomy, competence, and relatedness at work:
Putnick, D. L., and Bornstein, M. H. (2016). Measurement invariance conventions construction and initial validation of the work-related basic need satisfaction
and reporting: the state of the art and future directions for psychological scale. J. Occup. Organ. Psychol. 83, 981–1002. doi: 10.1348/096317909
research. Dev. Rev. 41, 71–90. doi: 10.1016/j.dr.2016.06.004 X481382
Ryan, R. M., and Deci, E. L. (2017). Self-Determination Theory: Autonomy and Vander Elst, T., Van den Broeck, A., De Witte, H., and De Cuyper, N. (2012).
Basic Psychological Needs in Human Motivation, Social Development, and The mediating role of frustration of psychological needs in the relationship
Wellness. New York, NY: Gilford. doi: 10.1521/978.14625/28806 between job insecurity and work-related well-being. Work Stress 26, 252–271.
Schaufeli, W. B., Bakker, A. B., and Salanova, M. (2006). The measurement of work doi: 10.1080/02678373.2012.703900
engagement with a short questionnaire: a cross-national study. Educ. Psychol. Vansteenkiste, M., and Ryan, R. M. (2013). On psychological growth and
Meas. 66, 701–716. doi: 10.1177/0013164405282471 vulnerability: basic psychological need satisfaction and need frustration as a
Schultz, P. P., Ryan, R. M., Niemiec, C. P., Legate, N., and Williams, G. unifying principle. J. Psychother. Integr. 23, 263–280. doi: 10.1037/a0032359
C. (2015). Mindfulness, work climate, and psychological need satisfaction Vansteenkiste, M., Ryan, R. M., and Soenens, B. (2020). Basic psychological need
in employee well-being. Mindfulness 6, 971–985. doi: 10.1007/s12671-014- theory: Advancements, critical themes, and future directions. Motiv. Emot. 44,
0338-7 1–31. doi: 10.1007/s11031-019-09818-1
Silva, M. N., Sánchez-Oliva, D., Brunet, J., Williams, G. C., Teixeira, P. J., White, R. W. (1959). Motivation reconsidered: the concept of competence. Psychol.
and Palmeira, A. L. (2017). “What goes around comes around”: antecedents, Rev. 66, 297–333. doi: 10.1037/h0040934
mediators, and consequences of controlling vs. need-supportive motivational
strategies used by exercise professionals. Ann. Behav. Med. 51, 707–717. Conflict of Interest: The authors declare that the research was conducted in the
doi: 10.1007/s12160-017-9894-0 absence of any commercial or financial relationships that could be construed as a
Tafvelin, S., and Stenling, A. (2018). Development and initial validation of the potential conflict of interest.
need satisfaction and need support at work scales: a validity-focused approach.
Scand. J. Work Organ. Psychol. 3, 1–14. doi: 10.16993/sjwop.30 Copyright © 2021 Olafsen, Halvari and Frølund. This is an open-access article
Teo, T. S., Srivastava, S. C., and Jiang, L. (2008). Trust and electronic distributed under the terms of the Creative Commons Attribution License (CC BY).
government success: an empirical study. J. Manage. Informat. Syst. 25, 99–132. The use, distribution or reproduction in other forums is permitted, provided the
doi: 10.2753/MIS0742-1222250303 original author(s) and the copyright owner(s) are credited and that the original
Trépanier, S. G., Fernet, C., and Austin, S. (2016). Longitudinal relationships publication in this journal is cited, in accordance with accepted academic practice.
between workplace bullying, basic psychological needs, and employee No use, distribution or reproduction is permitted which does not comply with these
functioning: a simultaneous investigation of psychological need terms.

Frontiers in Psychology | www.frontiersin.org 12 July 2021 | Volume 12 | Article 697306