Introduction

The Australian Curriculum, Assessment and Reporting Authority (ACARA) is the independent statutory authority responsible for development, implementation and analysis of Australia’s school National Assessment Program-Literacy and Numeracy (NAPLAN) which covers Years 3, 5, 7 and 9. NAPLAN provides comparative data between and within schools over years to allow tracking of student and school development (ACARA 2015), and these comparative data can be of use for policy makers, school administrators, teachers and parents. A focal point of the NAPLAN program is the publication of school results on the MySchool website (http://www.myschool.edu.au/). The purpose of the website is to provide school results in an accessible form to the Australian public to foster transparency and accountability of the Australian school system (ACARA 2015).

Since inception in 2008, NAPLAN and MySchool have received ongoing criticism from Australian scholars (For reviews see: Harris et al. 2013; Klenowski and Wyatt-Smith 2012; Polesel et al. 2012). Criticisms mirror those that have been raised by scholars investigating standardised testing regimes in other countries (Au 2011; Brockmeier et al. 2014; Herman and Golan 1993; Segool et al. 2013). Specifically, NAPLAN has been suggested to foster an unhelpful competitive culture between schools that results in narrowing of the curriculum (Hardy 2015; Klenowski and Wyatt-Smith 2012; Polesel et al. 2014; Thompson and Harbaugh 2013). It has also been suggested that NAPLAN has an overall negative impact upon teacher, parent and student well-being (Dulfer et al. 2012; Polesel et al. 2012; Wyn et al. 2014). We have been unable to find any studies that have specifically measured the level and extent of potential distress associated with NAPLAN, and the study reported in this paper was therefore designed to examine self-reported distress in a volunteer sample of teachers, parents and students associated with the independent schools sector in response to NAPLAN testing in Years 3 and 5.

NAPLAN and teacher well-being

The stakeholders who have received the most research attention regarding experiences of NAPLAN are teachers. A number of surveys and qualitative investigations have investigated teacher and principal perspectives regarding the impact of NAPLAN on pedagogy and any potential negative impact on their students (APPA 2013; Dulfer et al. 2012; IEUA 2010; Polesel et al. 2014; Thompson 2013; Thompson and Harbaugh 2013; Ward 2012; Wyn et al. 2014). In a survey of over 8000 Australian teachers (primary, secondary and principals), the majority of respondents reported an impact on teaching practices similar to those reported overseas (Dulfer et al. 2012; Polesel et al. 2014). NAPLAN has influenced curriculum to become more NAPLAN-oriented (i.e. literacy and numeracy focused) at the expense of other subject areas (e.g. Society and Environment, The Arts, and Physical Education). Additionally, NAPLAN has influenced how literacy and numeracy are taught (APPA 2013; Ward 2012). That is, in order to maximise NAPLAN performance, literacy and numeracy are taught in a structured way that closely aligns with the test, which has been referred to as “teaching to the test” (Polesel et al. 2014, p. 643; Thompson and Harbaugh 2013, p. 303). It has been argued that constraining the creativity and flexibility of teachers in this way may have a negative impact on their job satisfaction (Berliner 2011; Thompson and Cook 2014).

Additionally, increased pressure on teachers to produce strong NAPLAN results may negatively impact their own well-being. The increased pressure has been argued to largely stem from the publication of school results on the MySchool website (Dulfer et al. 2012; Harris et al. 2013; Klenowski and Wyatt-Smith 2012; Polesel et al. 2014; Ragusa and Bousfield 2015; Thompson 2013). In their large-scale survey, Dulfer et al. (2012) reported that around 90 % of educators believed weaker than expected NAPLAN results would have a negative impact upon the school’s reputation, parental perception of the school, ability to attract and retain students and staff morale. From a survey of over 800 educators from Western and South Australia, Thompson (2013) reported that in response to an open-ended question asking about the perceived negative impacts of NAPLAN, 44 % of the educators mentioned increased stress and pressure across teachers, parents and students as an issue. Similarly, a survey of over 1000 primary school principals found that many respondents indicated that NAPLAN can raise teacher stress levels in the lead up to the tests (APPA 2013). Prior research has therefore identified that NAPLAN testing has the potential to be a demoralising and stressful experience for teachers. However, no studies have studied the level and extent of distress experienced by teachers during NAPLAN testing compared with non-testing periods. An unanswered question is: Does the stress experienced by teachers during NAPLAN surpass the usual stressors associated with being a teacher? The present research aims to address this gap in the literature by examining teacher self-reported emotional distress during and after NAPLAN testing.

NAPLAN and parent well-being

The stakeholders who have received the least attention are parents. Existing research investigating parent perceptions consists of Australian Senate enquiries inviting submissions in 2010 and 2013 (Bousfield and Ragusa 2014), a survey by Australian market research company Newspoll commissioned by the Whitlam Institute (Newspoll 2013) and interviews with 26 parents by the Whitlam institute (Wyn et al. 2014). This research has been primarily focused upon parent perceptions regarding the usefulness of NAPLAN and perceived impact upon student well-being. Overall, this research suggests that two out of three parents have a relatively positive attitude towards the testing, and half of the parents report perceiving their child experiencing some level of stress associated with the testing. No research has directly investigated the level and extent of emotional distress experienced by parents in response to the testing. From transcript quotes provided by Wyn et al. (2014), there is evidence to suggest that some parents may experience elevated stress due to concern about their child’s performance and/or how their child will react to the testing. The present research aims to address this gap in the literature by examining parent self-reported emotional distress during and after NAPLAN testing.

NAPLAN and student well-being

As reviewed earlier, a few studies have asked teachers their impressions of the impact NAPLAN has on students (APPA 2013; Dulfer et al. 2012; Thompson 2013). From these studies, the precise extent of the impact of NAPLAN upon students is very unclear. For example, the survey of Dulfer et al. (2012) found not only that 90 % of teachers reported at least some students feeling stressed but also that 40 % reported some students looking forward to the testing. As another example, the primary principal survey found that two-thirds of respondents reported a belief that NAPLAN has a slightly negative impact upon students (APPA 2013).

To date, there are three studies that have examined actual student perceptions of NAPLAN (Belcastro and Boon 2012; Howell, 2012; Wyn et al. 2014). Belcastro and Boon’s (2012) study focused on student motivation rather than well-being, so will not be discussed in any detail here. Howell (2012) asked 100 students across Years 3, 5 and 7 in two schools to draw a picture about their NAPLAN experience and then qualitatively evaluated the themes present in the drawings. The authors reported that drawings were “overwhelmingly negative” (Howell 2012, p. 9). However, a closer inspection of Howell’s (2012, p. 10) results indicates that this pattern was only evident for one of the two schools, with the second school containing more balanced experiences.

Wyn et al. (2014) conducted interviews with 70 students evenly split across Years 5, 7 and 9, that were sampled across 16 schools. Most of the student interviews were carried out in groups. The section of the report of Wyn et al. (2014) dedicated to student perceptions of well-being consisted entirely of negative student anecdotes but then stated in the final paragraph of the section that “The majority of students reported that they did not like NAPLAN, but advised that they generally got through it without too many issues…” (Wyn et al. 2014, p. 27). In their executive summary, a conclusion made was that “…NAPLAN is a significant pedagogical intervention which has some positive uses, but is plagued by negative impacts on learning and on student well-being” (Wyn et al. 2014, p. 6). Therefore, both Howell (2012) and Wyn et al. (2014) appear to make strong negative conclusions based on mixed negative, neutral and positive findings.

In the Australian literature to date, the prevailing view appears to be that NAPLAN has a negative impact upon student well-being despite little evidence regarding the precise level and extent of distress experienced by students during testing (for example, Harris et al. 2013; Howell 2012; O’Keefe 2011; Polesel et al. 2012; Thompson 2013; Wyn et al. 2014). A further aim of the present study is to examine self-reported emotional distress of students during and after testing in order to contribute to the research literature.

The present study

As mentioned above, there is reason to believe that NAPLAN testing may have a negative impact across multiple stakeholders (i.e. educators, parents and students). However, no prior studies have quantified any impact. The primary aim of the present research is to explore this by examining teacher, parent and student self-reported emotional distress, in Years 3 and 5, across eleven independent schools in Western Australia. Based on prior research, we expected some negative impact to be experienced by all stakeholders; however, the size of effect was uncertain. NAPLAN is a test of performance, and like any test of performance (e.g. other school tests, ballet recital, job interview, first date, etc.), a slight increase in stress is an expected and a functional response. A long-standing psychological principle is the Yerkes-Dodson law that describes the inverted U shaped relationship between stress and performance (Cassady and Johnson 2002; Lowe et al. 2008; Salehi et al. 2010; Yerkes and Dodson 1908). That is, a slight elevation in stress is deemed adaptive in contexts of evaluation as a slight increase in arousal can facilitate concentration for the task at hand. However, after a certain optimal point, that is determined by both dispositional and situational factors, further stress can impair concentration and therefore diminish performance. Stress that is maintained at a high level for a prolonged period can exhaust physical and mental energy, resulting in diminished well-being of the individual. We therefore approach interpretation of our results with this in mind.

In our study, we use the same measure of distress across all stakeholders in order to enable direct comparisons amongst the groups. We therefore are able to directly investigate the previously unexplored questions: Who is the most affected by the NAPLAN testing—students, parents or teachers? And, what is the level and extent of distress associated with NAPLAN testing for each group? Respondents of the primary principals survey generally believed that Year 3 students were more negatively affected by testing compared with Year 5 students (APPA 2013). However, Howell (2012) found that Year 7 students produced a higher proportion of negative drawings compared to Year 3 and 5 students. Therefore, we were also interested in comparing across Years 3 and 5, in addition to the overall comparison across students, parents and teachers.

Surveys of teachers have revealed an existing belief that students may become more stressed as they are affected by their parents stress level, if the parent places a great deal of importance in performance and communicates (either verbally or non-verbally) their anxiety to their child (Thompson 2013; Wyn et al. 2014). In our study, we are able to explore if any positive association exists between parent distress and that of their child. We are also able to explore whether there is any positive association between teacher distress during testing and the overall distress experienced in their class of students.

Additionally, we also asked parents and teachers to provide estimates of the impact that NAPLAN testing has upon the well-being of all stakeholder groups of interest for this study (i.e. students, parents and teachers). It has previously been found that parents with a more negative attitude towards NAPLAN typically reported perceiving their child as experiencing higher levels of stress during testing (Newspoll 2013). Therefore, we expected to find a positive relationship between self-reported self-distress and perceived distress in others.

Method

Participants

Eighteen-member schools of the Association of Independent Schools of Western Australia were contacted by the research team to request participation in our study approximately two months prior to NAPLAN testing. Eleven school principals agreed to participate. Most of the schools declining participation were from rural areas. We acknowledge that our results are limited by a potential self-selection bias of schools that all hold student well-being as a high priority, and are likely not to be representative of schools generally, nor for all independent schools. As may be expected, all participating schools were above the median level (1000) of socio-educational advantage as determined by the Index of Community Socio-educational Advantage (ICSEA) that is published on the MySchool website, values ranging from 1051 to 1182 (Mean = 1148).

Participants consisted of 196 Year 3 students (aged 7–8 years old; 58.2 % female), 269 Year 5 students (aged 9–10 years old; 58.7 % female), 346 parents (Mean age = 43 years; 92.2 % female) and 40 teachers (Mean age = 37 years; 82.5 % female; 45 % Year 3 and 55 % Year 5) across 11 independent metropolitan schools in Western Australia. A specific break down of participants across schools is provided in Table 1. Prior to commencement of the study, institutional ethics approval was obtained from the Edith Cowan University ethics committee.

Table 1 The number of participants from different stakeholder groups across all schools involved in this study, split by gender
Full size table

Procedure

There are a number of established measures that have been designed to assess student test anxiety (for example, Lowe et al. 2011; Lowe et al. 2008; Sarason et al. 1958; Wren and Benson 2004). However, these measures could not be used to directly compare responses across different stakeholders or were sufficiently brief enough to minimise inconvenience upon participating schools. We therefore devised a new simple brief measure of emotional distress in order to achieve our purposes. This measure assessed emotional distress by asking the participant to rate their experience of six adjectives listed in the following order: Happy, worried, calm, sad, confident and afraid, on a 4-point scale (not at all, a little bit, quite a bit, a lot). The response scale for students had accompanying basic cartoon faces to increase engagement and clarity. For a copy of the survey instrument, please contact the first author.

Prior to commencement of the study, a lengthy pre-testing process was conducted trialling different versions of the measure with university students, parents, children and teachers in order to finalise the specific adjectives and response scale used. We examined the validity of this final measure by administering our brief measure to 126 first-year university students from Edith Cowan University (33 % 18–25 years, 16 % 26–30 years, 51 % over 30 years old; 84 % female), in addition to a well-established measure—The 21-item version of the Depression Anxiety Stress Scales (DASS-21) (Lovibond and Lovibond 1995). The DASS-21 is a commonly used measure of psychological distress with adults containing three 7-item subscales (depression, anxiety and stress), which is often summed to provide an overall distress score. When providing responses for both measures, the students reflected on their life during the month prior to the start of the university semester. The reliability and validity of the DASS-21 has been well established (Crawford et al. 2011; Crawford and Henry 2003). We therefore wished to compare our new brief scale of emotional distress with the well-established DASS-21. Inter-correlations and scale reliability coefficients are presented in Table 2. Our brief emotional distress scale achieved good internal consistency (α = .84) and good convergent validity with the overall DASS-21 (r = .78, p < .01). We acknowledge, however, that this validation of the final measure is limited, in that it was not performed on primary school-aged children.

Table 2 Inter-correlations between DASS-21 and the new brief emotional distress scale designed for this particular study. Cronbach’s alpha values are also included
Full size table

Students with parental consent were surveyed during normal classroom time within a 3-week period after NAPLAN testing. Members of our research team visited each school at an organised time and administered paper surveys to students by reading each statement aloud in an orderly fashion. Any student misunderstandings that arose were dealt with immediately before progressing further with the survey. Students filled out our brief well-being measure three times. Students completed the scale when asked “When doing NAPLAN maths testing I felt”, then “When doing NAPLAN reading testing I felt…” and finally “Since the NAPLAN testing week I have felt”. Additionally students were asked an open-ended question by providing a large box on the page with the instruction “In the box below please write or draw anything you think about NAPLAN”.

Consenting parents and teachers could elect to fill out a paper (with reply paid envelope), be telephoned or complete an online version of the survey at their convenience over a 6-week period after NAPLAN testing finished. Adult participants completed the brief emotional distress scale twice, asked “During the time my child/students participated in NAPLAN testing this year I felt” and “Since the NAPLAN testing week I have felt”. In Table 3, we provide the Cronbach’s alpha statistics for the measure as it was used in our study to provide evidence to support adequate reliability of the measure in the school context. Additionally, adult participants were asked “Do you believe that the pressure to do well on NAPLAN tests has a negative impact upon the well-being of: (students, parents and teachers)”. Participants rated their belief for each judgement on a 6-point scale (not at all, slightly, somewhat, moderately, very much, extremely).

Table 3 Cronbach’s alpha values for the brief emotional distress scale used in the present study
Full size table

Results

Self-reported student, parent and teacher emotional distress during and after NAPLAN

Students reported more distress during NAPLAN maths (M = 1.87, SD = .62), compared to NAPLAN reading (M = 1.77, SD = .67); however, this effect was very small, t(464) = 3.91, p < .01, r 2 = .03, and therefore, for further analysis, student reading and maths distress was averaged to produce a single measure. Furthermore, a moderate positive correlation was found between maths and reading distress (r = .63, p < .01).

Of primary interest was the difference between emotional distress during and after NAPLAN across different stakeholders, see Table 4. First, we checked if there was any large variation amongst schools for self-reported distress during NAPLAN by conducting a series of one-way ANOVAs that revealed no significant effect of school for Year 3 students, Year 5 students, parents or teachers (all Fs ≤ 1.27, ps > .05). Considering the fairly homogenous nature of our sample (i.e. relatively wealthy independent schools all willing to take part in a study focused on NAPLAN testing), the consistency across schools was not surprising. To examine emotional distress during and after NAPLAN across different stakeholders a 4 × 2 mixed design factorial ANOVA was conducted treating group (Year 3 students, Year 5 students, parents and teachers) as a between-participants factor and time (during and after) as a within-participants factor. Both the main effect of group (F(3, 847) = 5.54, p < .05, η 2 p  = .02) and time (F(1, 847) = 164.13, p < .05, η 2 p  = .02) were found to be statistically significant, in addition to the interaction between group and time (F(3, 847) = 11.80, p < . 05, η 2 p  = .04). To further understand the overall interaction effect, we conducted follow-up Bonferroni adjusted t-tests selectively based on questions of most interest. Following Field’s (2009) instructions, we applied the Bonferroni adjustment to the accepted p value for statistical significance by dividing by the number of comparisons carried out (.05/10 = .005). This post hoc procedure is a conservative approach that reduces family-wise error.

Table 4 Mean emotional distress for different stakeholder groups during and after NAPLAN, with percentage of responses falling within different score brackets for the measure
Full size table

As expected, compared to non-testing time, distress was found to be significantly higher during NAPLAN for all stakeholders: Year 3 students (t(195) = 6.90, p < .005, r 2 = .20), Year 5 students (t(268) = 13.26, p < .005, r 2 = .40), parents (t(345) = 7.30, p < .005, r 2 = .13) and teachers (t(39) = 4.08, p < .005, r 2 = .30). Follow-up comparisons across all stakeholder groups for emotional distress during NAPLAN only revealed that Year 3 students reported a lower level of distress compared with Year 5 students (t(463) = 4.41, p < .005, r 2 = .04) and teachers (t(234) = 3.14, p < .005, r 2 = .04), with the other comparisons failing to reach significance. Note that although there are significant differences between groups, effect sizes are very small indicating that despite some slight differences, emotional distress during NAPLAN was fairly similar across the stakeholder groups.

An important finding was that despite increased distress during NAPLAN testing, no mean value for any of the stakeholder groups was higher than “a little bit” on our brief emotional distress scale. Furthermore, the percentage of participants across all stakeholder groups that scored in the “quite a bit” to “a lot” range was small, both during and after NAPLAN, see Table 4. When compensating for typical distress levels as indicated by ‘after NAPLAN’ responses, results suggest NAPLAN produced a severe negative reaction in approximately 3 % of students, parents and teachers in our sample. Although this remains concerning for those 3 %, the results show that in the present sample, 97 % did not report a severe negative reaction to NAPLAN. The results therefore confirmed that NAPLAN elevates distress. The level of distress reported during NAPLAN by the present sample, however, did not appear to be severe for the majority of the respondents. In fact, levels of distress appeared less than what one might reasonably expect to be associated with a testing experience such as NAPLAN.

In addition to quantifying distress levels, we were interested in exploring if parent and teacher distress associated with distress reported by students. A significant correlation was found between parent and student self-reported emotional distress during NAPLAN, although the association was small (r = .18, p < .01, n = 297). This relationship remained nearly identical when splitting by year group (Year 3 r = .19 & Year 5 r = .20). Each teacher was allocated a ‘classroom distress’ score by averaging across student emotional distress scores during NAPLAN for each teacher. No relationship was observed between classroom distress scores and teacher self-reported emotional distress during testing (r = .06, p = .72, n = 38).

Student open-ended responses

As part of our survey, students were asked to write or draw in an open-ended fashion their feelings towards NAPLAN. An open-ended response was provided by 82 % of the student sample. Some example student responses are presented in Fig. 1.

Fig. 1
figure 1

Examples of a positive, b neutral and c negative open-ended responses from Year 3 students

Full size image

Student open-ended responses were coded as positive, neutral or negative by the first author and a research assistant. Inter-rater agreement was high (Cohen’s Kappa = .84). Any disagreements were discussed and resolved to produce the percentages presented in Fig. 2. A significant difference was found between the Year 3 and Year 5 pattern of responses (χ 2(2) = 24.61, p < .01, Cramer’s V = .25). A higher proportion of Year 3 students produced positive responses compared to Year 5 students. The majority of Year 3 responses were positive, and Year 5 student responses were fairly evenly distributed across positive, neutral and negative categories. Consistent with our emotional distress findings, student responses to the open-ended question did not reveal any prevailing negativity towards NAPLAN testing. This consistency affords greater confidence in our findings.

Fig. 2
figure 2

Percentage of open-ended responses coded as positive, neutral and negative for Year 3 (N = 196) and 5 (N = 269) students

Full size image

Parent and teacher estimates of the impact of NAPLAN upon the well-being of other stakeholders

The extent that parents and teachers believe the pressure to do well in NAPLAN affects others is presented in Table 5. For both parents and teachers, the average response fell between “somewhat” to “moderately”, with the exception that parents believed NAPLAN impacted parents “slightly” to “somewhat”. Overall, these results are consistent with the emotional distress results. In our study, the experience of NAPLAN was not overly stressful for most parents and teachers and additionally was not believed to have a severe impact on others. However, there was a fairly consistent response rate across estimations that ~25 % of respondents reported the negative impact of NAPLAN in the “very much” to “extremely” range. Therefore, whilst most respondents did not perceive a large impact, there was still a substantial proportion holding such a belief.

Table 5 Mean parent (N = 346) and teacher (N = 40) perceived negative impact of NAPLAN upon the well-being of students, parents and teachers, with frequency percentages for each of the response categories
Full size table

We expected that the self-experience of NAPLAN, as measured by our brief emotional distress scale, would be positively associated with estimations of how others experience NAPLAN. Results supported this hypothesis. Parent self-reported emotional distress during NAPLAN was found to positively correlate with estimations of the impact of NAPLAN upon the well-being of students (r = .43), parents (r = .40) and teachers (r = .30), all ps ≤ .01. Likewise, teacher self-reported emotional distress during NAPLAN was also positively correlated with their estimations of the impact of NAPLAN upon the well-being of students (r = .50), parents (r = .34) and teachers (r = .53), all ps ≤ .01. Therefore, results suggest that compared with a teacher or parent experiencing lower levels of distress during NAPLAN, a teacher or parent experiencing higher levels of distress during NAPLAN typically also assumes that NAPLAN causes more distress in others.

Discussion

The present research examined student, parent and teacher self-reported emotional distress during and after NAPLAN in a sample of 11 independent schools in Western Australia. Across students, parents and teachers, we found an increase in distress levels during NAPLAN compared with after the testing. However, the average level of distress during testing did not exceed “a little bit” for all groups on our self-report distress measure. It is also important to recognise that our data suggest NAPLAN produces a severe negative reaction (i.e. “quite a bit” to “a lot” of distress) in approximately 3 % of the sampled students, parents and teachers, after accounting for some individuals having pre-existing levels of high distress. We caution that the sample of schools were atypical of the general sector and furthermore may also be atypical of the independent schools sector in the way these schools approach NAPLAN testing. These limitations suggest that the percentage of students, parents and teachers reporting distress in the broader schooling context could potentially be higher than what we have reported, and it is important that further studies be undertaken to estimate the prevalence in the Australian population. It is important that the percentage of those who report high distress be managed, and that schools enact strategies to minimise any severe reaction to the testing. There is an opportunity for schools to monitor reactions of students during NAPLAN for early identification of test anxiety problems and enact intervention where appropriate.

In the present study, overwhelming negativity in student drawings or statements was not observed in open-ended responses from students. Year 3 students provided a higher proportion of positive responses, and Year 5 students were evenly split across negative, neutral and positive responses. Year 5 students also reported more distress during the testing compared to Year 3 students. This finding is inconsistent with beliefs reported in a survey of primary school principals that Year 3 students are more impacted by the testing than older children (APPA 2013). It is beyond the scope of our present study to determine the precise reasons behind why the Year 5 students reported a more negative experience of NAPLAN compared to the Year 3 students. Perhaps, it is simply a result of being in the school system for longer and developing a more nuanced appreciation of what testing means for the student. This is however only speculation and understanding how the experience of repeated testing like NAPLAN changes over time for students is an avenue for future research.

In our review of the literature, we noted that a lack of research studies investigating the effect of NAPLAN on stakeholders currently limits arguments about NAPLAN’s impact. Prior studies consist of educator impressions regarding the impact upon students (APPA 2013; Dulfer et al. 2012; Thompson 2013) and two qualitative investigations of actual student experiences (Howell 2012; Wyn et al. 2014). Our findings add to the existing literature by providing a more direct assessment of the level and extent of emotional distress during testing compared with what has previously been examined, across all stakeholders. The present study is the first time as far as we are aware that the level and extent of distress associated with NAPLAN has actually been measured, and the disparity between our results and those found in the extant literature is quite marked.

Interpretation of our findings is, however, limited to the context of our sample (i.e. independent schools with above average socio-educational advantage (ICSEA) scores). Principals of these schools all reported that enrichment of socio-emotional well-being was a school priority, and citation of specific well-being programmes to achieve this aim was common. NAPLAN was largely considered just another test within a battery of school assessments that does not hold any special consequences for the students, teachers or school as a whole. From a survey of West and South Australian teachers, results suggested that the (negative) impact of NAPLAN upon pedagogy may be more prevalent in state schools and schools situated in lower socio-economic status (SES) areas (Thompson and Harbaugh 2013). More research is required to ascertain whether our findings can be replicated in government schools. It may be that in general, greater importance is placed upon NAPLAN achievement by principals in government schools. This could produce a more pressure-laden school climate during testing that impacts the experience of students. On the other hand, research has reported that Australian children from higher SES backgrounds may have greater expectations placed upon them for academic success by their parents (Dandy and Nettelbeck 2002). Therefore, it could be reasoned that more parental pressure to perform in higher SES contexts may place some students at greater risk for experiencing distress during testing. Further research is required to understand the interplay between different school cultures and the NAPLAN experience of stakeholders within those cultures.

A methodological limitation of our research was how participants were asked to retrospectively think back upon their NAPLAN experience some time after the testing (it varied from 1 to 3 weeks after testing). We concede that relying upon the memory of the experience instead of asking at the time of testing introduces a level of inaccuracy into our data. Furthermore, by asking participants to reflect upon their memory of their emotional experience during testing, and then subsequently asking about their emotional experience since the testing, may have led participants to contrast the two situations that could act to potentially amplify differences. For example, a participant may believe that they should have been feeling more distressed during NAPLAN compared to afterwards, and answered in a way consistent with that belief. For the present research, we took a cautious approach, as we were concerned about surveying students during the testing time as we did not want to risk elevating their distress levels. Additionally, we did not wish to impose upon the schools during a time (i.e. during NAPLAN) they were already experiencing a disruption to their usual routines. All of these issues need to be taken under consideration for future studies that aim to measure distress associated with testing. A further limitation is the convergent validation of our brief emotional distress scale with the DASS-21, using a university student sample. It will be important to further validate the scale with other child-oriented measures of emotional distress in the future.

In our sample, no relationship was found between teacher distress during NAPLAN, and the distress of children from their class. A small significant relationship (r = .18) was found between parent distress and child distress during NAPLAN. It is beyond the scope of our study to determine if this relationship is more indicative of a student responding to their parent’s distress, or parents responding to distress of their child. The overall low levels of distress observed in our study may be masking a stronger association, and future research is required to investigate the interplay between parent and child emotional response during periods of evaluation for the child. We found stronger positive (moderate) relationships between parent and teacher emotional distress during testing with their perceptions regarding how much NAPLAN was negatively impacting the well-being of others. Our findings indicate that parents and teachers reporting more distress during testing tended to also perceive more distress in others. This is consistent with a long-standing psychological principle known as the false consensus effect that refers to a tendency to over-estimate the extent that others share our attitudes, beliefs and emotional experiences (Ross et al. 1977; Van Boven and Loewenstein 2003). A similar finding to our own is that parents with a more negative attitude towards NAPLAN have been reported to report more distress experienced by their child (Newspoll 2013).

NAPLAN testing has had many harsh critics. One major criticism has been to question the impact NAPLAN has upon stakeholder well-being. Our research provides evidence to demonstrate that NAPLAN does not have any major impact upon well-being for the majority of stakeholders in certain contexts. More evidence is required to fully understand the experience of NAPLAN across states and territories, urban and rural, and low/high SES contexts.