Systematic Review and Meta-Analysis of Positive Psychology Interventions in Workplace Settings

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Overall, this manuscript is well done. The authors do a good job addressing the purported gap in the literature - IGLO levels of intervention effectiveness in organizational settings differentiating subjective vs psychological well-being in face to face as well as remote delivery formats over time while accounting for moderating variables. Great job meticulously comparing the effect sizes (both within-groups and between-groups).

I just have a few comments on some minor issues I think are worth attending to:

On Line 119-120, you note "a research assistant who independently reviewed the work using an AI-based tool." Could you please give a little more detail on what exactly this "tool" was and what it did?

On Line 191-193, you note "The subjective performance measures included self-reported ratings of job contribution, task performance, interpersonal support, energy levels at work, and organizational virtues. These measures provide a holistic view of performance". Since, in the overall paragraph, you mention that you're only including subjective measurements (i.e. self-report measures), I would strongly recommend you add the qualifier "somewhat holistic view of performance" (since some objective dimensions of this measurement are lacking).

Line 80 - "mega-analysis" should be "meta-analysis"

Line 587-588 - "Intervention format was also relevant programs delivered 587 in person tended to show stronger results than those implemented remotely." <-- awkward grammar that needs rewording.

On page 16, SWB - Western group k = 6, g =-0.87. Even though the CI is [-1.87, 0.13], some explanation in discussion would help as to conjecture why the overall effect was negative here for Western groups.

Overall though, I think that this manuscript is worth publishing and would contribute to the overall discussion on Positive Psychology Interventions in the workplace with nuances explored in this meta analysis.

Author Response

Thank you for your comments

All changes compared to the original version are shown in bold and green.

Comment: On Line 119–120, you note “a research assistant who independently reviewed the work using an AI-based tool.” Could you please give a little more detail on what exactly this “tool” was and what it did?

Response: We clarified this sentence in Section 2 by specifying the tool used (DeepSeek, 2024 version) and describing its limited role as a support resource to classify interventions as single- or multi-component based on predefined criteria. The sentence now reads:

"Data extraction and study selection was conducted by the first author with help from a research assistant who checked it independently and made use of an AI-based software (DeepSeek, 2024) as a supporting resource..."

Comment: Line 191–193: Please add the qualifier “somewhat holistic view of performance.”

Response: The text was updated as suggested. The phrase now reads:

"...These measures provide a somewhat holistic view of performance..."

Comment: Line 80 – "mega-analysis" should be "meta-analysis"

Response: We respectfully clarify that the term “mega-analysis” is intentionally used to refer to Carr et al. (2023), who themselves define their work as a "mega-analysis" in the title. Therefore, although we removed one instance to avoid redundancy, we preserved this term where it is cited as part of Carr’s work.

Comment: Line 587–588 – Grammar revision needed for sentence on intervention format.

Response: Sentence reworded for clarity. It now reads:

“Format of intervention also turned out to be a significant factor: in-person programs generated better results than programs that were carried out through remote interventions.”

Comment: The Western group for SWB has g = –0.87; discussion should offer an explanation.

Response: We apologize for the confusion. We discovered that there was a mix-up in the moderator table. The value g = –0.87 actually refers to the non-Western group and is already discussed and justified in the manuscript. Additionally, the previous version of the moderator table mistakenly showed the within-group results instead of the correct between-group comparisons, which are the ones relevant for moderator analysis. We have corrected this by recalculating and carefully reviewing the moderator results for the performance variable using the appropriate between-group data. We also revisited the original studies to verify the accuracy of the moderator coding. The updated version now reflects the correct values and interpretations.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

General comments:

• This paper is well-written, the logic is easy to follow, and the desired contribution is well-defined. However, theoretical/conceptual aspects could be improved/clarified (some suggestions have been provided in the review below).
• Concrete, practical implications (e.g. what does this mean for employers who want to use such interventions? what should / shouldn't they use them for?). Sometimes are implied or mentioned abstractly in passing but not really discussed.
• Future research directions are lacking. A few simple ones could be added either to the discussion or the conclusion. I have also suggested a few throughout my review.
• Consider replacing certain more casual terms by more scientific terms ones. e.g.
  • "looked at"; lines 58, 69, 91) - "explored" or "examined."
  • "zoomed in on" line 484
• In-text citations might need to be revised. For example,
  • there are at least two articles that could be "BMJ" but one is cited BMJ and the other is cited using the author's name. Revise for consistency. (eg. line 226)
  • when citing software (eg Jamovi. (line 198)), I believe it is customary to cite the version and year, and to add it to the reference list.
  • line 519-522 is there a source that could support this claim? perhaps a handbook or guide?
• Headings, subheadings and some formatting to be revised.

Specific comments and recommendations by section:

1. Introduction:
   • The article starts with a mention of job stress, burnout and the deficit-based/clinical approach and suggests that a positive-oriented approach may be pertinent.
     • This is true, and very much in line with the positive psychology movement. However, the article then goes on to measure SWB and PWB, which are positive WB measures. It may be pertinent to bridge this gap with one or two sentences to explain why positive interventions and WB measures may be pertinent in the context of stress and burnout, because this may not be obvious for readers less familiar with positive psychology. Mentioning Keyes (2002) dual continuum model might be helpful here.
     • Alternatively, because the mention of stress/burnout is only to introduce the topic, perhaps the article could be introduced in another way; for example, by discussing the increased attention that employers have been paying to employee wellbeing.
   • Citing Lyubomirsky for cultivation of positive emotions (...) is appropriate, but adding a proper definition and citation for PPIs is also recommended. For example, see Donaldson et al. (2019) and/or the Positive Work and Organizations (PWO) literature generally.
   • Some concepts are defined without adding a proper citation (e.g. Hedonic and eudaimonic well-being)
   • Knowledge gap is clearly presented. However, sometimes the justification for filling this gap is not given (or it is simply implied). For example, one of the main differences between this paper and Donaldson (2019) is that this paper distinguishes between SWB and PWB outcomes. Why is this important? What theoretical and/or practical implications could this have?  
   • lines 87-89 akward phrasing.
     • I suggest: "Although they didn't note any moderating effect of gender on PPI effects, Carr et al. (2020; 2023) deemed cultural setting to matter more.
     • also: more than what?.
   • I think it was an interesting idea to categorize interventions using the IGLO model. Later on, the paper shows that I G and L level interventions are sparce. However, other fields of study (e.g. burnout, JD-R) have shown time and time again that organizational interventions are needed to benefit worker health (e.g. primary intervention). This might suggest that further research is needed to see whether this is also true for positive psychology interventions.
   • Aim of paper is clear, consistent with the literature review, and the paper does what it sets out to do. However, the title states this is a systematic review AND a meta-analysis; but section 1.1 only says Aim of this REVIEW. Perhaps the editor can help clarify if this paper should be considered one or the other or both.
2. Material and method
   • I do not know what the generally agreed-upon way to report using AI tools is. However, for the sake of transparency and replicability, it may be advisable to provide more information. For example, what AI tool was used? How was it used? (e.g. what prompts or criteria or formulas?). 
   • I believe it might be common practice to provide the search strings in the text (e.g. in the format of a table) instead of at the end of the document in an appendix (Editor may confirm or disconfirm based on the journal's preferred format).
   • In terms of theoretical grounding, one or two sentences clarifying how the research terms were chosen would be helpful. This could include basing the search terms on those used by previous reviews for example. (there is an example in the Donaldson 2019 reference previously cited)
   • Search strategy: line 124 mentions PsycINFO but line 142 says PsycArticles.
     • which is correct?
     • If PsycArticles was used, why was PsychINFO not used? 
3. Results
   • Sometimes when formulas or tests are mentioned, the author's name is cited but not the year. I would add the year (and the references in the reference list if that is not already the case).
   • Table 2: country column, the meaning of "EEUU" as a country is unclear. (This might be obvious to readership in some parts of the world but not others?) I suggest adding it to the notes.
   • If possible, presenting the larger tables 2, 3 and 6 in landscape format might help with readability.
   • lines 337-344 there is a discussion of bias, but no mention of what kinds of bias there could be. (see for example the previously-cited Carolan reference)
   • Figures 2 and 3 might require light proofreading. For example: after "et al", some sources have nothing, some have a period only, and some have a period and a comma. 
   • generally speaking, when dicussing performance, it might be more appropriate to specify "self-reported" performance, since the measures are not based on "objective" metrics. Later this could be cited as a limit and an avenue for future research.
   • For most of the paper, the "DV's" are presented in the same order (SWB, PWB, performance). However, in some of the results subsections, the order changes (eg. subsection 3.5). If possible, consider revising to present the results in the same order for fluidity (consistency).
   • lines 401-403 avoid introducing new concepts without definition and/or justification. I suggest removing "collectivistic" and "individualistic," unless the author discusses this later. For example, when interpreting results, these features about societies could influence the results of PPIs... how? why? (especially given the individual focus of PPIs).
4. Discussion
   • well structured, easy to follow.
   • lines 470-472 reminds the reader of the contribution with regards to examining PWB and WB distinctly. Why is this pertinent/important? What do individual studies show about the difference between these two? How might different interventions influence one or the other differently? (e.g. adding meaningful tasks might increase PWB more than SWB? what does this mean for practice? for work in general? (from a social science perspective, given the journal))
   • lines 565-567 the concepts of optimal functioning and thriving are introduced. These ideas could be introduced earlier (ie. in the introduction, instead of stress/burnout). Also, consider referencing the notion of flourishing, as it is central to positive psychology. In other words, I think this is a good insight that could be developed further.
   • consider discussing the limits of using subjective performance measures (beyond the fact that they are self-reported)

Author Response

Thank you for your comments

All changes compared to the original version are shown in bold and green.

General Comments

Comment:
The reviewer suggested that theoretical and conceptual aspects could be clarified; that practical implications should be more explicitly discussed; that future research directions are lacking; and that terminology and citation consistency should be improved.

Response:
We fully agree and have implemented the following changes:

• Theoretical clarifications have been added in both the Introduction and Discussion to better articulate why examining psychological well-being (PWB) and subjective well-being (SWB) separately is conceptually and practically relevant in workplace contexts. This includes references to Keyes' (2002) dual continuum model and the differentiation in how interventions may target hedonic versus eudaimonic dimensions of well-being.
• Practical implications are now more explicitly addressed in the Conclusion, particularly highlighting that in-person interventions tended to show stronger results and offering insights on what organizations may consider when implementing PPIs.
• Future research directions have been added to the end of the Discussion section, recommending further studies at the group, leader, and organizational levels of the IGLO framework, and also advocating for the standardization of objective performance measures across studies to enable better comparison between sectors.
• Informal phrases such as “looked at” and “zoomed in on” have been replaced with more scientific terms like “examined” and “focused on”.
• All citations have been reviewed for consistency. The manuscript now follows the author–date format throughout. Software such as Jamovi is now cited with version and year (Jamovi, version 2.4, 2023) and included in the reference list.
• Regarding lines 519–522: Thank you for pointing this out. We acknowledge that line numbers may differ depending on formatting and platform. In our document, these lines read:
  “What is the impact of PPIs on performance? The meta-analysis yielded an overall Hedges’ g of around 0.42, reflecting an overall improvement in self-reported performance. Importantly, comparatively few workplace studies examined performance outcomes of PPIs. Donaldson et al. (2019) is the sole previous meta-analysis specifically on…”
  We appreciate the suggestion to support this section further. Although this statement is partially supported by Donaldson et al. (2019), we are happy to add further citations (e.g., Meyers et al., 2013; Schueller et al., 2014) if the editor considers it necessary.

1. Introduction

Comment:
The reviewer noted a need to clarify the relevance of positive well-being measures in relation to stress and burnout, to better define PPIs, and to improve citation and justification of key concepts such as SWB and PWB.

Response:

• We have added two sentences that connect the rationale for Positive Psychological Interventions (PPIs) to the prevention of workplace stress and burnout. This includes referencing the dual continuum model of Keyes (2002), which justifies the co-existence of mental illness and low well-being and supports the relevance of focusing on PWB and SWB as outcomes in stress-prone environments.
• Additionally, we included a sentence noting the growing attention that employers are giving to employee well-being, offering a practical rationale for studying PPIs in the workplace even if not all interventions are stress-focused.
• A more precise definition of PPIs is now provided in the introduction, citing both Donaldson et al. (2019) and the Positive Work and Organizations (PWO) literature to strengthen the conceptual foundation.
• The definitions of hedonic and eudaimonic well-being are now accompanied by proper citations (Ryan & Deci, 2001; Waterman, 1993; Koydemir et al., 2021).
• The importance of differentiating SWB and PWB has been more clearly justified. We now explain that these constructs may be differently affected by various types of interventions—e.g., strength-based or meaning-based interventions might impact PWB more, while gratitude exercises may increase SWB—and that their separation allows for more nuanced understanding and design of PPIs.
• The awkward phrasing previously found around lines 87–89 has been revised. The new sentence now reads:
  “Carr et al. (2020, 2023), although they did not find gender to significantly moderate the effects of PPIs, observed that cultural setting more frequently emerged as a significant moderating variable.”
• We acknowledge the limited presence of group-, leader-, and organization-level interventions in our dataset. Based on this, we now advocate for further research targeting these underexplored levels of the IGLO model—especially given that other fields such as job design or burnout prevention have shown that interventions at the organizational level often produce longer-lasting effects.
• It is changed “Aim of this review” for “aym of this research”.

 

 

2. Materials and Methods

Comment: I do not know what the generally agreed-upon way to report using AI tools is. However, for the sake of transparency and replicability, it may be advisable to provide more information. For example, what AI tool was used? How was it used? (e.g. what prompts or criteria or formulas?).

Response:
We thank the reviewer for this observation. In the revised version of the manuscript (Section 2), we have added a more detailed explanation regarding the use of AI tools. Specifically, we now state that the research assistant used DeepSeek (2024 version) as a support tool for classifying interventions as single- or multi-component, based on the operational definitions provided in Section 2.3. This use was strictly complementary and subject to verification and consensus by the research team. The revised sentence now reads:
"Data extraction and study selection were conducted by the first author, with support from a research assistant who independently reviewed the work and used an AI-based tool (DeepSeek, 2024 version) solely as a support mechanism to classify interventions as single- or multi-component, using the operational definitions provided in Section 2.3. Discrepancies were resolved by discussion with the co-authors."

 

Comment: I believe it might be common practice to provide the search strings in the text (e.g. in the format of a table) instead of at the end of the document in an appendix.

Response:
We appreciate this recommendation. In response, we have moved the complete Boolean search strategy from the appendix to the end of Section 2.1 in the main text to improve accessibility and transparency. The format has also been slightly adapted for clarity while preserving its original structure.

 

Comment: In terms of theoretical grounding, one or two sentences clarifying how the research terms were chosen would be helpful. This could include basing the search terms on those used by previous reviews, for example Donaldson (2019).

Response:
Thank you for this suggestion. We have now included a sentence at the beginning of Section 2.1 explaining that the selection of search terms was informed by prior meta-analyses, particularly Donaldson et al. (2019) and Hendriks et al. (2020), to ensure consistency with prior operationalizations of PPIs and well-being constructs.

 

Comment: Search strategy: line 124 mentions PsycINFO but line 142 says PsycArticles. Which is correct? If PsycArticles was used, why was PsychINFO not used?

Response:
We thank the reviewer for spotting this inconsistency. It was a typographical error: the correct database used was PsycArticles, we excluded any publication different from articles. This has been corrected in line and verified throughout the manuscript to ensure consistency.

 

3. Results

Comment:
Sometimes when formulas or tests are mentioned, the author's name is cited but not the year.
Response:
We appreciate this observation. We carefully reviewed all statistical tests and formulas throughout the manuscript and ensured that author names are consistently accompanied by the corresponding publication year (e.g., Hedges & Olkin, 1985; Egger et al., 1997; Rosenthal, 1979). The references were added or corrected where needed.

 

Comment:
Table 2: The meaning of "EEUU" as a country is unclear.
Response:
We have explaned “EEUU” in notes.

 

Comment:
If possible, present the larger Tables 2, 3 and 6 in landscape format to improve readability.
Response:
We agree that clarity is important. We reviewed the formatting options available but were unable to implement a full landscape layout due to submission platform constraints. Nevertheless, we revised the layout, countries names and spacing of the affected tables to enhance their readability

 

Comment:
Lines 337–344: There is a discussion of bias, but no mention of the types of bias.
Response:
We believe the reviewer is referring specifically to the discussion of publication bias. While we had already included funnel plots and Egger/Begg tests in the results, we now specify that the main focus of bias analysis was publication bias. We chose not to expand on other types (e.g., social desirability, confirmation bias) because they were outside the scope of the statistical meta-analysis, but we do acknowledge this as a limitation in the discussion.

 

Comment:
Figures 2 and 3 might require light proofreading; inconsistencies appear in the formatting after “et al.”
Response:
We reviewed Figures 2 and 3 and standardized the formatting of all author citations. Punctuation following “et al.” has been corrected to ensure APA consistency, and minor formatting issues were revised for clarity.

 

Comment:
When discussing performance, it would be more appropriate to specify “self-reported performance.”
Response:
This is an excellent point. We now consistently use the term “self-reported performance” throughout the manuscript to more accurately reflect the nature of the outcome measures. This has also been included as a limitation in the discussion, where we recommend future research to incorporate more objective performance indicators.

 

Comment:
The DV’s (SWB, PWB, performance) should follow a consistent order throughout the results sections.
Response:
We agree with this suggestion. All subsections in the Results section have been reviewed and revised to consistently present outcome variables in the order: subjective well-being (SWB), psychological well-being (PWB), and performance.

 

Comment:
Lines 401–403: Avoid introducing the terms “collectivistic” and “individualistic” without explanation.
Response:
Following your recommendation, we have removed the terms “collectivistic” and “individualistic” and replaced them with a more general explanation of cultural context differences. This avoids introducing undefined constructs and keeps the interpretatio more accessible, while maintaining consistency with referenced studies such as Hendriks et al. (2020) and Koydemir et al. (2021).

4. Discussion

Comment:
Lines 470–472 remind the reader of the contribution regarding the separate analysis of PWB and SWB. Why is this important? What do individual studies show about the difference between these two? How might different interventions influence one or the other differently?
Response:
Thank you for this valuable insight. We have expanded the discussion to clarify why distinguishing between psychological well-being (PWB) and subjective well-being (SWB) is both theoretically and practically relevant. Specifically, we now note that while SWB tends to reflect short-term emotional states and general life satisfaction (Diener, 1984), PWB is more deeply rooted in purpose, autonomy, and self-realization (Ryff, 1989). This distinction is important because some interventions—such as gratitude journaling or mindfulness—may influence affective states (SWB), whereas interventions focused on strengths or meaning (e.g., job crafting, values clarification) may primarily impact PWB. We have also emphasized the practical implications of this: tailoring interventions based on whether the goal is to increase hedonic or eudaimonic well-being may improve their effectiveness in workplace settings.

 

Comment:
Lines 565–567 introduce the concepts of optimal functioning and thriving. These ideas could be introduced earlier in the manuscript, and the concept of flourishing could also be considered, as it is central to positive psychology.
Response:
We appreciate this observation. While we opted to retain the structure of the introduction, we now include earlier references to the concepts of optimal functioning and flourishing. Specifically, we introduce the notion of flourishing (Keyes, 2002; Seligman, 2011) as a guiding framework for PPIs in the workplace. In the discussion, we further develop how PPIs are aligned with promoting flourishing by enhancing meaning, engagement, and positive relationships at work—key domains in the PERMA model (Seligman, 2018). These additions reinforce the theoretical basis for using PPIs beyond symptom reduction.

 

Comment:
Consider discussing the limitations of using subjective performance measures (beyond the fact that they are self-reported).
Response:
This suggestion is well taken. We have expanded the limitations section to address the broader implications of relying solely on subjective performance measures. In particular, we mention that self-reports can be influenced by social desirability bias, lack of benchmarking, and participants’ own understanding of performance standards. While subjective performance ratings offer valuable insight into perceived functioning and motivation, future research would benefit from integrating objective indicators (e.g., productivity metrics, supervisor evaluations) to triangulate these findings.

 

Author Response File: Author Response.docx