pith. sign in

arxiv: 2604.05209 · v1 · submitted 2026-04-06 · 💻 cs.SE

Corporate Training in Brazilian Software Engineering: A Qualitative Study of Useful Learning Experiences

Rodrigo Siqueira , Antonio Oliveira , Breno Alves de Andrade , Lidiane C S Gomes , Danilo Monteiro Ribeiro This is my paper

Pith reviewed 2026-05-10 18:43 UTC · model grok-4.3

classification 💻 cs.SE
keywords software engineeringcorporate trainingqualitative studythematic analysislearning experiencescontinuous learningBrazil
0
0 comments X

The pith

Brazilian software professionals perceive training as useful when it aligns directly with daily work demands and delivers immediate applicability.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This qualitative study investigates what kinds of learning experiences Brazilian software engineering professionals consider genuinely useful in corporate settings. Open-ended responses from 195 participants were examined through thematic analysis, revealing five recurring dimensions of valued experiences. The analysis shows that usefulness centers on practical relevance rather than any single format, with technical updating and hands-on application emerging as dominant and mutually reinforcing. Organizations can use these insights to design training that combines multiple modalities instead of relying on isolated events.

Core claim

Five themes emerged from the data: Continuous Technical Updating (T1), Practical and Applied Learning (T2), Formal Academic Education (T3), Social Learning and Networking (T4), and Leadership Development and Soft Skills (T5). Technical updating and practical application dominate in both frequency and co-occurrence. Perceived usefulness is strongly tied to alignment with daily work demands and immediate applicability. Formal academic education and practical learning are viewed as complementary rather than competing, and genuinely valued experiences span technical, academic, social, and self-directed dimensions, underscoring the imperative of continuous learning in software engineering.

What carries the argument

Thematic analysis of open-ended responses from 195 professionals, supported by frequency counts, lemmatization, and co-occurrence analysis between themes.

If this is right

  • Organizations should design integrated training ecosystems that combine technical, academic, social, and soft-skills elements rather than isolated events.
  • The strong convergence of technical updating and practical application indicates that continuous learning must remain a priority in software engineering roles.
  • Formal academic education complements practical learning and should not be treated as a substitute for on-the-job applicability.
  • Valued learning includes social networking and leadership development as additional dimensions beyond pure technical content.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Training platforms could incorporate real-time applicability checks to increase the share of experiences rated as useful.
  • Similar qualitative studies in other countries could test whether the dominance of practical alignment is specific to the Brazilian context or more general.
  • Self-directed learning resources might gain value if they explicitly link content to participants' current work tasks.

Load-bearing premise

The self-reported perceptions from the 195 professionals accurately reflect genuinely useful learning experiences, and the thematic analysis captured all relevant dimensions without significant sampling or interpretive bias.

What would settle it

A study that measures objective job performance or skill retention after different training types and finds that experiences rated low on usefulness produce equal or better outcomes than those rated high would challenge the central claim.

Figures

Figures reproduced from arXiv: 2604.05209 by Antonio Oliveira, Breno Alves de Andrade, Danilo Monteiro Ribeiro, Lidiane C S Gomes, Rodrigo Siqueira.

Figure 1
Figure 1. Figure 1: Phases of thematic analysis and quantitative refine [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Co-occurrence graph between identified themes. [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
read the original abstract

Context: Quantitative studies can identify statistical predictors of training quality, but they often fail to capture what professionals themselves consider genuinely useful learning experiences and why. Objective: This study qualitatively investigates which types of learning experiences are perceived as most useful by Brazilian software engineering professionals and what characteristics define this usefulness. Method: Open-ended responses from 195 software engineering professionals were analyzed using Thematic Analysis, supported by frequency and lemmatization analysis using IRAMUTEQ and co-occurrence analysis between themes. Results: Five themes emerged: Continuous Technical Updating (T1), Practical and Applied Learning (T2), Formal Academic Education (T3), Social Learning and Networking (T4), and Leadership Development and Soft Skills (T5). Technical updating and practical application dominate professionals' accounts. Formal education, social learning, and soft skills are also valued as complementary dimensions. Conclusions: Perceived usefulness is strongly tied to alignment with daily work demands and immediate applicability. The convergence of technical updating (T1) and practical application (T2) in both frequency and co-occurrence reinforces the imperative of continuous learning in software engineering. Useful learning is not reducible to a single modality: genuinely valued experiences span technical, academic, social, and self-directed dimensions. Formal academic education and practical learning are perceived as complementary rather than competing. Organizations should design training ecosystems that integrate these dimensions rather than delivering isolated events.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 3 minor

Summary. The paper claims that a thematic analysis of 195 open-ended responses from Brazilian software engineering professionals reveals five key themes of useful learning experiences: Continuous Technical Updating (T1), Practical and Applied Learning (T2), Formal Academic Education (T3), Social Learning and Networking (T4), and Leadership Development and Soft Skills (T5). It concludes that perceived usefulness is strongly linked to alignment with daily work demands and immediate applicability, with T1 and T2 dominating in frequency and co-occurrence, and that useful learning spans multiple complementary dimensions rather than a single modality.

Significance. If the results hold, this study offers important implications for corporate training in software engineering by emphasizing the need for integrated training ecosystems that combine technical updating, practical application, formal education, social networking, and soft skills development. It underscores the value of continuous learning tailored to immediate work needs, which could guide organizations in designing more effective professional development programs, particularly in emerging markets like Brazil.

major comments (3)
  1. [Methods] Methods section: The thematic analysis follows Braun & Clarke steps and uses IRAMUTEQ for frequency/co-occurrence, but reports no inter-coder reliability metrics (e.g., Cohen's kappa), saturation evidence, or member checking. This directly undermines confidence in the emergence and dominance of T1/T2 as the central claim, since theme construction from self-reports is inherently interpretive.
  2. [Methods] Methods section: Sample recruitment details are absent (e.g., whether convenience sampling via LinkedIn/professional networks was used and how selection/response bias was mitigated). Without this, the link between the 195 responses and the claim that usefulness is 'strongly tied to alignment with daily work demands' cannot be fully assessed for representativeness.
  3. [Results] Results section: Frequency and co-occurrence findings are summarized but lack specific tables, exact counts, or visualizations showing how T1 and T2 converge to 'reinforce the imperative of continuous learning.' This makes the quantitative support for the qualitative conclusions difficult to verify.
minor comments (3)
  1. [Abstract] Abstract: The conclusions paragraph could briefly note the reliance on self-reported perceptions to balance the strong claims about 'genuinely valued experiences.'
  2. [Discussion] Discussion: Include additional participant quotes for each of the five themes to increase transparency of how themes were derived from the data.
  3. Ensure consistent abbreviation of themes as T1–T5 across all sections, tables, and figures.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed feedback on our manuscript. The comments identify important opportunities to strengthen transparency and verifiability. We address each major comment below and indicate the revisions that will be incorporated in the next version of the paper.

read point-by-point responses

  1. Referee: [Methods] Methods section: The thematic analysis follows Braun & Clarke steps and uses IRAMUTEQ for frequency/co-occurrence, but reports no inter-coder reliability metrics (e.g., Cohen's kappa), saturation evidence, or member checking. This directly undermines confidence in the emergence and dominance of T1/T2 as the central claim, since theme construction from self-reports is inherently interpretive.

    Authors: We acknowledge that the current Methods section provides limited detail on the interpretive process. Our analysis followed Braun and Clarke's reflexive thematic analysis, in which themes are actively constructed rather than discovered through consensus coding; consequently, traditional inter-coder reliability statistics such as Cohen's kappa are not routinely reported in this tradition. Nevertheless, we agree that greater transparency is warranted. In the revised manuscript we will expand the Methods section with a step-by-step description of the coding iterations, explicit criteria used to determine thematic saturation (no new themes after a defined number of additional responses), and a clear statement that member checking was not feasible given the anonymous survey format. These additions will allow readers to better evaluate the robustness of T1 and T2. revision: partial

  2. Referee: [Methods] Methods section: Sample recruitment details are absent (e.g., whether convenience sampling via LinkedIn/professional networks was used and how selection/response bias was mitigated). Without this, the link between the 195 responses and the claim that usefulness is 'strongly tied to alignment with daily work demands' cannot be fully assessed for representativeness.

    Authors: The referee is correct that recruitment procedures receive only brief mention. The 195 responses were obtained through an online survey distributed via Brazilian software-engineering professional groups and LinkedIn communities. In the revised version we will add a dedicated subsection describing the sampling frame, inclusion criteria, invitation channels, and the steps taken to encourage participation across different experience levels and company sizes. We will also expand the Limitations section to discuss the implications of convenience sampling and self-selection bias for the generalizability of the observed link between usefulness and daily-work alignment. revision: yes

  3. Referee: [Results] Results section: Frequency and co-occurrence findings are summarized but lack specific tables, exact counts, or visualizations showing how T1 and T2 converge to 'reinforce the imperative of continuous learning.' This makes the quantitative support for the qualitative conclusions difficult to verify.

    Authors: We agree that the current presentation of frequency and co-occurrence data is too summarized. The revised Results section will include a table reporting the exact number and percentage of responses coded to each theme, a co-occurrence matrix (or network diagram) highlighting the intersections between T1 and T2, and a brief narrative that directly ties these counts to the claim that their convergence reinforces the importance of continuous learning. These additions will make the quantitative support for the qualitative conclusions fully verifiable. revision: yes

Circularity Check

0 steps flagged

No circularity: qualitative analysis derives directly from survey data

full rationale

The paper is a qualitative study that applies thematic analysis (Braun & Clarke) to open-ended responses from 195 professionals, supported by IRAMUTEQ frequency and co-occurrence counts. No mathematical derivations, equations, fitted parameters, predictions, or self-citations serve as load-bearing steps. All claims about theme dominance (T1/T2) and usefulness characteristics are extracted from the collected data without reduction to prior fitted quantities or self-referential definitions. The derivation chain is self-contained and non-circular.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

This is an empirical qualitative study relying on standard social-science methods and self-reported data. No free parameters are fitted, no new entities are postulated, and the only background assumptions are domain-standard ones about the validity of thematic analysis for perception data.

axioms (1)
  • domain assumption Thematic analysis is an appropriate and reliable method for identifying patterns in open-ended responses about learning experiences.
    Invoked when the abstract states that responses were analyzed using Thematic Analysis supported by frequency and co-occurrence tools.

pith-pipeline@v0.9.0 · 5555 in / 1484 out tokens · 54281 ms · 2026-05-10T18:43:01.581650+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. It's Not About Whom You Train: An Analysis of Corporate Education in Software Engineering

    cs.SE 2026-04 unverdicted novelty 4.0

    Survey of 282 professionals shows training mandatoriness dominates perceptions of corporate SE education quality, with minimal influence from sociodemographic variables.

Reference graph

Works this paper leans on

13 extracted references · 13 canonical work pages · cited by 1 Pith paper

  1. [1]

    1977.Social learning theory

    Albert Bandura and Richard H Walters. 1977.Social learning theory. Vol. 1. Prentice-hall Englewood Cliffs, NJ

    work page 1977

  2. [2]

    Paul D Coverstone. 2003. IT training assessment and evaluation: A case study. In Proceedings of the 4th conference on Information technology curriculum. 206–215

    work page 2003

  3. [3]

    Breno Alves de Andrade, Rodrigo Siqueira, Lidiane C. S. Gomes, Antonio Oliveira, and Danilo Monteiro Ribeiro. 2025. A Mapping Study About Training in Industry Context in Software Engineering. InProceedings of the XXXIX Brazilian Sympo- sium on Software Engineering (SBES ’25) (CBSoft 2025). SBC, Recife, PE, Brazil, To appear

    work page 2025

  4. [4]

    Sarv Devaraj and S Ramesh Babu. 2004. How to measure the relationship between training and job performance.Commun. ACM47, 5 (2004), 62–67

    work page 2004

  5. [5]

    Wellynton Diniz, Marcela Valença, César França, Alessandro Santos, and Mariana Pincovsky. 2024. The Skill Gap in Software Industry: A Mapping Study. In Simpósio Brasileiro de Engenharia de Software (SBES). SBC, SBES, BRA, 192–200

    work page 2024

  6. [6]

    Alan Dix and L Uden. 2000. Lifelong Learning for Software Engineers. InIFIP World Computer Congress (ICEUT 2000)

    work page 2000

  7. [7]

    Jeffrey D Facteau, Gregory H Dobbins, Joyce EA Russell, Robert T Ladd, and Jeffrey D Kudisch. 1995. The influence of general perceptions of the training environment on pretraining motivation and perceived training transfer.Journal of management21, 1 (1995), 1–25

    work page 1995

  8. [8]

    Pierre Ratinaud. 2009. IRAMUTEQ: Interface de R pour les Analyses Multidimen- sionnelles de Textes et de Questionnaires [Computer software]

    work page 2009

  9. [9]

    Eduardo Salas and Janis A Cannon-Bowers. 2001. The science of training: A decade of progress.Annual review of psychology52, 1 (2001), 471–499

    work page 2001

  10. [10]

    Eduardo Salas, Scott I Tannenbaum, Kurt Kraiger, and Kimberly A Smith-Jentsch

    work page

  11. [11]

    The science of training and development in organizations: What matters in practice.Psychological science in the public interest13, 2 (2012), 74–101

    work page 2012

  12. [12]

    Amalia Santos and Mark Stuart. 2003. Employee perceptions and their influence on training effectiveness.Human resource management journal13, 1 (2003), 27–45

    work page 2003

  13. [13]

    Gareth Terry, Nikki Hayfield, Victoria Clarke, Virginia Braun, et al. 2017. The- matic analysis.The SAGE handbook of qualitative research in psychology2, 17-37 (2017), 25

    work page 2017