Is there a fit between PhD candidates’ and their supervisors’ perceptions on the impact of COVID-19 on doctoral education?

Doctoral supervisors’ and supervisees’ experiences of the impact of COVID-19

There is strong evidence that supervisory relationships play a key role in successful PhD degree completion (Lovitts, 2001; Peltonen et al., 2017; Pyhältö et al., 2015). Accordingly, supervisory relationships potentially offer a central resource for overcoming the challenges caused by the COVID-19 pandemic for PhD candidates. Yet, harnessing the potential and building a grounding for offering well-fitted supervisory support for the PhD candidates to overcome the challenges caused by the pandemic requires shared understanding of the influences. This cannot be taken for granted, but typically calls for shared sensemaking about the influences (Janssen et al., 2021).

Based on the restricted empirical evidence, it seems that the COVID-19 pandemic has predominantly had a negative impact on PhD candidates (Atkinson et al., 2021; Aubry et al., 2021; Camerlink et al., 2021; Donohue et al., 2021; Myers et al., 2020). For instance, we recently showed that PhD candidates typically estimated the COVID-19 pandemic as having a negative impact on their study progress and study well-being (Pyhältö et al., 2022). The negative influences reported by the candidates were related to impaired access to the data, erosion of scholarly support networks, reduced access to institutional resources, poor work–life balance and reduced well-being (Pyhältö et al., 2022). The destructive COVID-19 effects for PhD candidates have also been shown to include a decline in the weekly working hours devoted to research (Myers et al., 2020), delays in thesis submission (Stamp et al., 2021), reduced opportunities for networking (Guintivano et al., 2021) and professional development (Kappel et al., 2021), as well as reduced involvement and experienced usefulness of online events attended (Raby and Madden, 2021). There is also tentative evidence that frequency of supervision may have declined during the pandemic (Pyhältö et al., 2022). A reason for this might be fewer informal meetings such as coffee break meetings with the supervisors due to the lockdowns. In addition, reduced levels of well-being due to the COVID-19 pandemic have been detected (Atkinson et al., 2021; Donohue et al., 2021), including increased stress levels (Camerlink et al., 2021; Guintivano et al., 2021), mental and physical fatigue (Adarmouch et al., 2020; Stamp et al., 2021) and poorer work–life balance (Aubry et al., 2021).

It also appears that the impacts of the pandemic have not been evenly distributed across PhD candidates (Minello et al., 2020; Myers et al., 2020; Staniscuaski et al., 2021; Vincent-Lamarre et al., 2020), potentially contributing to further inequalities and hence varied support needs between them. For example, during the first 10 weeks of lockdown in the USA, women’s research productivity measured in the number of publications decreased by 14% relative to men (Cui et al., 2021). A similar decline has also been detected in Australia, China, Italy, Japan, The Netherlands and the UK (Cui et al., 2021; Guintivano et al., 2021). Moreover, we recently showed that international candidates, those at the mid phase of their studies, candidates studying full time, engaging in research teams and those from natural sciences had an increased risk of suffering from more severe COVID-19 influences compared to their peers (Pyhältö et al., 2022). The results imply that PhD candidates are likely to need a range of types of support, including but not limited to supervisory support, to overcome the hardship caused by the pandemic.

Empirical evidence on the impact of COVID-19 on doctoral supervisors and supervision is even more limited. Results on the few studies on COVID-19 influences on supervision are mixed: some studies have reported a decline in supervisors’ stress (Betts, 2021; Else, 2021), reduced productivity particularly among those with teaching duties (Myers et al., 2020; Watermeyer et al., 2021), blurring of boundaries between the personal and the professional (Lasater et al., 2021) and an increase in workload due to additional tasks and responsibilities that have resulted from the pandemic (Le, 2021). However, in other studies, no such effects or even opposite influences have been detected (Eringfeld, 2021). For instance, in our recent study on Finnish doctoral supervisors, two-thirds of the supervisors (66%) reported that their supervisory responsibilities had not changed because of the outbreak of COVID-19, whereas a quarter of the supervisors (24%) reported that the pandemic had increased their supervisory responsibilities, and 10% estimated that their supervisory responsibilities had declined (Löfström et al., submitted). Those reporting changes in supervision estimated that the challenges of supervisory responsibilities caused by COVID-19 were primarily related to recognising when doctoral candidates needed help and supporting the candidates (Jung et al., 2021). Similarly, Eringfeld (2021) showed that supervisors experienced reduced levels of performance demands and stress, and higher levels of well-being during the pandemic. Supervisors’ experiences on the impact of COVID-19 has also been reported as varying according to seniority, whether they supervise full-time or part-time candidates, and depending on the scholarly field (Gill et al., 2020; Lambrechts and Smith, 2020; Löfström et al., submitted).

Overall, the limited evidence on the impact of COVID-19 on doctoral supervisors and supervisees implies that they might have had varied experiences, and hence perceive influences of COVID-19 on PhD experience differently. Also, variation across the disciplines and study status (full-time versus part-time) might occur regarding the alignment between the supervisors’ and supervisees’ estimates of the pandemic influences, considering that the differences detected across the varied groups of PhD candidates. Yet, there have been no empirical studies on the topic.

Doctoral education at the research-intensive university in Finland

Finland has one of the highest rates of doctoral degree holders per capita (OECD, 2014), and has adopted a system nationwide graduate school relatively recently. At the research-intensive university, all doctoral candidates have belonged to one of the four doctoral schools, and to one of the university’s 33 doctoral programmes since 2014. Studying for a doctorate (a total of 240 ECTS) involves conducting doctoral research (80%; i.e. 200 ECTS) that is launched from the beginning, and complementary coursework (20%; i.e. 40 ECTS) based on a personal study plan (Andres et al., 2015). Doctoral candidates write their dissertation as either a monograph or as a set of articles. Most doctoral candidates write an article-based doctoral dissertation (Pyhältö et al., 2022), typically including three or four peer-reviewed published articles and a summary. Doctoral candidates are typically supervised by at least two supervisors, with the main supervisor usually being a full professor holding a permanent position at the university. The thesis examination proceeds in three stages: firstly, two external examiners appointed by the faculty pre-examine the dissertation; secondly, if approved, a public defence follows the pre-examination; and finally, the faculty grants the PhD degree. The target time for completing the doctorate is four years studying full time; however, the average time for degree completion is five or six years (Pyhältö et al., 2022). There are no tuition fees, but the universities, projects or foundations do not automatically provide funding for the doctoral candidate.

Participants

The participants in the study comprised 561 doctoral supervisors (275 women, 50%; 266 men, 49%; and 6 non-binary, 1%) and 768 PhD candidates (502 women, 67%; 234 men, 31%; and 18 non-binary, 2%) from a large multi-field research-intensive university in Finland. The response rate of the supervisors was 16%, and 17% for PhD candidates. The typical age of the participating PhD candidates was 30–34 years, and 40–44 years for supervisors. In terms of disciplinary distribution, the participants represented PhD candidate and doctoral supervisor populations in the case university well. Also, in terms of gender distribution, the supervisor data represented the population well, but women were slightly overrepresented in the PhD candidate data. Of the candidates, 62% reported studying full time and 38% part time. Similarly, about a third of the supervisors reported supervising part-time candidates (see Table 1).

Data

In this study, we used a large-scale quantitative approach with two separate data sets collected in 2021. The data sets were combined to compare the supervisors’ and PhD candidates’ perceptions of the impact of COVID-19 on candidates’ progress and well-being, and how the perceptions were related to supervisors’ and supervisees’ experiences of engagement, stress, exhaustion and cynicism in their work. The data from PhD candidates were collected between April and May 2021 using a modified version of the cross-cultural doctoral experience survey (C-DES) validated in previous studies (Pyhältö et al., 2016; see also C-DES manual: Pyhältö et al., 2018). The DES has been originally developed and validated among Finnish PhD candidates with various data sets since 2007. The survey was developed based on the series of qualitative studies and literature on the main determinants of doctoral experience (Stubb et al., 2014; Vekkaila et al., 2013, 2014). Since 2008, the survey has been validated with data collected from nine countries, and it has been translated into several languages (see details in the cross-country doctoral experience manual 2018: Pyhältö et al., 2018). The validation procedures have included translation back procedures, several pilots with thinking aloud procedure and series of confirmatory analysis. Also, researcher teams from each country have been involved in the survey validation to ensure proper contextual adaptation. The supervisory experience survey was first developed in 2010 by the first author and her team. The development and further adaptions involve piloting with supervisors, and series of quantitative analysis. Recently, cross-country data have been also collected using supervisory experience survey. The data from doctoral supervisors for this study were collected between August and September 2021 using a modified version of supervisory experience survey validated in previous studies (Pyhältö et al., 2015). Both data sets were collected via online surveys available in Finnish, Swedish and English. The links to the surveys were sent via e-mail to the participants by using the Doctoral Schools’ PhD candidates’ and supervisors’ mailing lists. All the participants were informed about the study before the data collection. No identifying information or incentives were used. Participation in the study was voluntary.

The following measures were used to examine the PhD candidates’ perceptions of COVID-19 and their study well-being:

• one item on the COVID-19 pandemic impact on PhD candidates’ study well-being scale;

• one item on the COVID-19 pandemic impact of doctoral study progress scale;

• one item stress scale (Elo et al., 2003);

• study burnout scale comprising exhaustion (five items) and cynicism scales (six items) (adapted from Schaufeli et al., 2002a); and

• study engagement scale (nine items) (adapted from Schaufeli et al., 2002b; Salmela-Aro and Upadaya, 2012).

For measuring doctoral supervisors’ perceptions of the same themes, the following scales were used:

• one item on the COVID-19 pandemic impact on PhD candidates’ study well-being scale;

• one item on the COVID-19 pandemic impact of doctoral study progress scale;

• one item stress scale (Elo et al., 2003);

• burnout scale comprising exhaustion (five items), cynicism (three items) and sense of inadequacy scales (three items) (Pyhältö et al., 2018 adapted from Maslach Burnout Inventory); and

• work engagement scale (nine items) (Schaufeli et al., 2006).

All the scales were one- to seven-point Likert scales. For the items, see Appendix.

In Finland, an ethics review is required when research involves intervention in the physical integrity of research participants; deviates from the principle of informed consent; involves participants under the age of 15 being studied without parental consent; exposes participants to exceptionally strong stimuli; risks causing long-term mental harm beyond that encountered in normal life; or signifies a security risk to subjects (Finnish National Board on Research Integrity, 2019). None of these conditions affected this study.

Analysis

To get the overall view of the PhD candidates’ and supervisors’ perceptions of the impact of the COVID-19 pandemic on candidates’ study progress and study well-being, the means and standard deviations were calculated. To further analyse the fit between the PhD candidates’ and supervisors’ perceptions, the independent samples t-test was used. Variation across the disciplines in PhD candidates’ and supervisors’ perceptions of the impact of COVID-19 were explored with one-way analysis of variance and differences based on candidates’ study status with the independent samples t-test. For pairwise comparisons of the disciplinary differences, Tukey’s HSD test was used. Independent samples t-test was also used to analyse differences in supervisors’ and candidates’ perceptions of impact of COVID-19 within the disciplines. Bonferroni correction was applied when multiple t-tests were conducted to same groups. Finally, to analyse the associations between the perceptions of the impact of COVID-19 on PhD candidates’ and supervisors’ well-being, Pearson’s correlation coefficients between the study variables were calculated.

PhD candidates’ and supervisors’ perceptions of the impact of COVID-19 on study progress and candidates’ well-being

Candidates reported that the pandemic had hindered their study progress (M = 4.12, SD = 2.14) and reduced their doctoral study related well-being (M = 4.39, SD = 2.01). Similarly, doctoral supervisors often estimated the pandemic as having a negative impact on doctoral candidates’ progress (M = 4.60, SD = 1.82) and well-being (M = 4.87, SD = 1.57). Further investigation showed that the supervisors estimated the negative impact of COVID-19 on the supervisees’ progress to be higher [t(1,274) = −4.34, p < 0.001] and the decrease in their study well-being to be more severe [t(1,292) = −4.82, p < 0.001] than the candidates themselves did. Accordingly, in general, the supervisors estimated the impact of COVID-19 on supervisees’ progress and well-being to have been more detrimental than the supervisees did.

Variation in PhD candidates’ and supervisors’ perceptions of the impact of COVID-19 across and within the disciplines and candidates’ study status

Some disciplinary differences in the impact of the COVID-19 pandemic were detected among the candidates [F(3, 740) = 4.96, p < 0.01]. The candidates in the natural sciences estimated that the COVID-19 pandemic had reduced their study-related well-being more severely than their peers in the health sciences (see Table 2). No such disciplinary differences in the estimated impact of COVID-19 on candidates’ well-being were detected among the supervisors. There were no disciplinary differences in PhD candidates’ estimates on COVID-19 impact on study progress. However, some disciplinary differences were detected among the supervisors: supervisors in health sciences estimated COVID-19 influences on study progress to be more detrimental than supervisors in humanities and social sciences [F(3, 463) = 2.770, p < 0.05].

Some disciplinary differences in the fit between candidates’ and supervisors’ perceptions of the impact of COVID-19 were also detected: In the health sciences, the supervisors assessed the impact of COVID-19 on the study progress [t(292) = −5.51, p < 0.001] and study well-being [t(306) = −4.03, p < 0.001] more negatively, than the supervisees themselves did. In the humanities and social sciences, the supervisors estimated the negative pandemic influence on the candidates’ well-being [t(438) = −2.93, p < 0.01] to be higher than the candidates themselves did, but, there was no difference between the candidates’ and the supervisors’ perceptions of the impact of COVID-19 on the study progress. In the natural sciences and in the environmental, food and biological sciences, no differences were detected in the supervisors’ and candidates’ perceptions of the impact of COVID-19.

The candidates who were conducting their doctoral research full time reported that the COVID-19 pandemic had both decreased their study-related well-being [t(580) = 8.56, p < 0.001] and hindered their study progress [t(554) = 5.55, p < 0.001] more often than those studying part time. Similarly, the doctoral supervisors, whose supervisees were typically studying full time, perceived more often that the pandemic had influenced negatively the progress [t(545) = 2.55, p < 0.05] and well-being of their candidates [t(307) = 4.92, p < 0.001] than supervisors supervising mostly part-time candidates.

When the differences in full-time candidates and the supervisors supervising mainly full-time candidates were explored, no differences in the perceptions of the impact of COVID-19 on study well-being or study progress were detected. However, the doctoral supervisors who supervised mainly part-time candidates reported significantly more often that the pandemic had influenced their supervisees’ progress [t(420) = −3.83, p < 0.001] and study well-being [t(414) = −4.34, p < 0.001] than the part-time candidates themselves reported.

Association between well-being and the perceptions of COVID-19 influence on candidates

The perceptions of negative impact of COVID-19 on the progress of the doctoral research and on study-related well-being were related to increased levels of stress, exhaustion and cynicism, and reduced levels of research engagement among the PhD candidates (see Table 3).

Further investigation showed that the supervisors’ perceptions of the negative impact of COVID-19 on candidates’ progress and well-being were also related to their own well-being (see Table 4). More specifically, the supervisors’ negative estimates of the impact of COVID-19 on their supervisees were related to experiencing elevated levels of stress, exhaustion, cynicism and inadequacy in their work. However, the perceptions of negative impact of COVID-19 on their supervisees were not related to work engagement among the supervisors.

Findings in the light of previous research

We set out to investigate the alignment between PhD candidates’ and supervisors’ estimates of the impact of COVID-19 on study progress and the candidates’ study well-being, and whether the estimates were related to supervisors’ and supervisees’ well-being. Results showed that, in general, the supervisors estimated that the impact of COVID-19 on supervisees’ progress and well-being had been more detrimental than the supervisees did. This implies that supervisors do recognise and are sensitive to the challenges faced by the PhD candidates resulting from the pandemic. Accordingly, although earlier studies suggest that the impact of the pandemic on the supervisors tended to be less severe and to some extent varied (Betts, 2021; Camerlink et al., 2021; Eringfeld, 2021) compared to supervisees, the supervisors’ estimates of the impact on supervisees matched well with those of the supervisees. Further investigation showed that supervisors’ and supervisees’ estimates about the impact of COVID-19 were also well aligned regarding study status: PhD candidates who studied full time considered the impact of COVID-19 to be more detrimental compared to the candidates studying part time, as did the supervisors who primarily supervised full-time candidates. In general, supervisors estimated that the impact of COVID-19 on candidates’ progress and well-being to be more negative than the candidates themselves did. This implies a certain degree of mismatch between the supervisees’ and supervisors’ estimates on the impact of COVID-19. A reason for this might be that because of their experience, supervisors are able to make more accurate estimations, for example, about the long-term consequences of data collection delays for the study progress. Moreover, supervisors may worry about the supervisees’ future study trajectories, and consider that providing extra-support to be their responsibility, even though they are also facing a pandemic crisis for the first time in their careers (Lasater et al., 2021). Some disciplinary differences regarding supervisors’ and supervisees’ estimates of the impact of COVID-19 were identified: while supervisors’ estimates of the impact in health sciences exceeded the estimates of the supervisees, in humanities and social sciences such a difference was only detected regarding candidates’ well-being. In natural sciences and in environmental, food and bio sciences, no such differences between the supervisors’ and supervisees’ estimates about COVID-19 were detected, implying a better understanding of the impact. A reason for this might be that the number of full-time students was highest within these disciplines, which potentially allows more frequent interaction and hence shared sensemaking about COVID-19 between the supervisors and supervisees, which is reflected in a better alignment between the perceived impact.

Both supervisors’ and supervisees’ negative estimates of the impact of COVID-19 on PhD candidates were related to reduced levels of well-being: increased levels of stress and burnout symptoms. In this regard, our results support the findings of previous studies suggesting that COVID-19 pandemic increases stress and the risk of developing burnout symptoms among both the supervisees and supervisors (Atkinson et al., 2021; Donohue et al., 2021; Betts, 2021; Camerlink et al., 2021). The results further extend the findings of previous studies by showing that hardships caused by the pandemic for PhD candidates reflect negatively on both supervisors’ and supervisees’ well-being. The finding implies that as an intensive long-term professional relationship, supervisory interaction is challenged by the pandemic, resulting in a decline in supervisors’ and supervisees’ well-being. However, while the reported negative influences were related to the reduced levels of research engagement among the PhD candidates, no such relationship were detected regarding supervisors’ work engagement. A reason for this may be that though potentially aligned, supervisors’ research is not dependent on their supervisees’ research and their work is more varied compared to the PhD candidates, whose primary task is to promote their doctoral research. On the contrary, the findings may indicate that the core resource of supervisors’ work contributing to their work engagement does not stem from supervisory relationship, and hence are not affected by the problems faced in this domain. Moreover, because of being less experienced in research and having more limited access to institutional resources, the supervisees might be more emotionally vulnerable and less resilient in facing hardships caused by the pandemic. Accordingly, our paper contributes to the new understanding on interrelation between the supervisors’ and supervisees’ well-being under particular circumstance, i.e. during the COVID-19 pandemic. However, the results concerning the interrelation between supervisors and supervisees have potentially more general implications: firstly, supervisors’ well-being is likely to reflect on the kind of supervisory practices applied, and secondly, the experienced quality of supervisory relationship is likely to have impact not only on supervisee’s well-being but also on supervisor’s occupational well-being. The hypothesis is supported by our recent results on supervisors’ well-being suggesting that the supervisors’ perceptions of quality of supervisory relationship was related to higher levels of work engagement, and decreased levels of burnout symptoms, especially cynicism. The satisfaction with supervisory competences was also positively associated with work engagement and negatively with stress and burnout symptoms. Furthermore, receiving support from the scholarly community was related to increase in work engagement, and decrease in stress and burnout symptom (Tikkanen et al., forthcoming). Similarly, a recent study on UK research supervisors showed that despite a high proportion reporting supervisory responsibilities, only half of the supervisors suggested their workplace formally recognised their supervision in workload allocation models, and that a third of them have experienced stress caused by the concern related to supervision that have kept them awake at night over the past 12 months (UK Council for Graduate Education, 2021). However, further studies are needed to test the hypothesis.

Implications

Our results suggested that COVID-19 pandemic has had a negative impact on the PhD candidates’ well-being and progress across the disciplines and that both supervisors and supervisees recognise it. This calls for developing effective means of support to overcome the hardships to avoid long-term negative consequences for the candidates’ degree completion, career trajectories and the future of the academy. Because supervision is repeatedly shown to be one of the main determinants of doctoral experience and degree completion (Cornér et al., 2017; Ives and Rowley, 2005; Lovitts, 2001; Peltonen et al., 2017; Pyhältö et al., 2012, 2015), supervisors’ understanding of the impact of the COVID-19 pandemic on the candidates provides a reason for providing support for the candidates to stay resilient in facing the challenges and overcome the hardships. Accordingly, the fact that supervisors acknowledged the impacts and were sensitive to them implies that supervisors are in general well tuned with the supervisees’ experiences. This would suggest that the shared sensemaking is needed more in co-creating means to overcome the challenges rather than just building an understanding of whether the PhD candidates are influenced by the pandemic or not. However, it is important to note that measures taken to overcome the hardships cannot be limited only to supervision, but calls for efforts at all levels of doctoral education ranging from policies and funding instruments to institutional support tools and individual actions. Our results further suggest that the negative COVID-19 influences are related to an increased risk of suffering from stress and burnout symptoms among both the supervisors and supervisees. This suggests that the pandemic can compromise the well-being of both supervisors and supervisees via having a negative impact on the latter. Accordingly, actions on overcoming the influences should consider the supervisors’ support needs that would allow them to be responsive to their supervisees’ needs, without compromising their own occupational well-being.

Methodological limitations

The study has some methodological strengths and limitations that should be considered in drawing a conclusion. A major strength of the study is that it explored the fit in supervisors’ and supervisees’ views of the COVID-19 pandemic with comparable measures by using data collected from both groups from the same institution. The response rates for both groups of participants were somewhat low. However, regardless of the low response rate, in terms of age and disciplinary distribution, the PhD candidates were a good representation of the whole PhD candidate population at the case university. Women were slightly over presented in the data. The distribution of men and women was a good representation of doctoral supervisors at the university, and they were from all four doctoral schools. However, we were not able to match the supervisors with their own supervisees because we wanted to protect the anonymity of the respondents. Therefore, one should be cautious in making conclusions about the supervisory relationship between the participants. Instead, the results reflect the fit in the perceptions concerning the impact of COVID-19 at the more general institutional level. Measures used in the study can be considered reliable. We used a cross-sectional design in this study, and hence, causal conclusions or conclusions about development of the perceptions cannot be drawn.