From Game Records to Training Action
Symbolic Performance Informatics in Collegiate Chess Self-Analysis
Abstract
Chess produces unusually detailed records of cognitive performance: move sequences, positions, annotations, engine evaluations, and traces of error. Yet the availability of rich game data does not automatically produce learning. This qualitative interpretive case study examines how ten collegiate chess players in the University of the Philippines Diliman chess context use game records as personal informatics resources for perceived development. Drawing on semi-structured interviews and artifact-elicitation prompts, the study analyzes how players capture, stabilize, annotate, mediate, judge, and act on their own game data. The findings show that collegiate chess self-analysis operates as a form of symbolic performance informatics: a self-tracking process in which symbolic records of cognitive and strategic performance are converted into self-knowledge and training action. Players preserved games through notation sheets, memory-based reconstruction, PGN files, platform histories, and Lichess studies; transformed records through replay, comments, symbolic marks, and critical-position labels; and judged development through both numerical indicators, such as ratings and fewer blunders, and experiential indicators, such as confidence, calmness, preparedness, and pattern recognition. Digital tools including Lichess, Chess.com, ChessBase, Analyze This, CT-ART, and Stockfish expanded access to evaluation, but participants repeatedly emphasized that engine output did not automatically become understanding. Recorded moves and best-move lines required interpretive mediation through prior experience, coaches, peers, databases, books, and self-regulatory judgment. The study contributes to personal informatics by extending the field beyond bodily and sensor-generated tracking, identifies interpretive mediation as the theoretical hinge between reflection and action, and positions personal chess archives as learning infrastructure whose value depends on capture, replayability, annotation, retrieval, and repair under collegiate constraints.
Keywords
symbolic performance informatics, personal informatics, chess self-analysis, personal archives, interpretive mediation, self-regulated learning, human-AI learning, information behavior
Introduction
Chess produces records that make performance unusually available for later scrutiny. A completed game is not only an outcome; it is a sequence of decisions that can be replayed, annotated, compared, and interpreted after the event. Through notation sheets, digital archives, PGN files, engine analysis, databases, and online replay platforms, players can return to their own performances in fine detail. These records preserve traces of judgment, error, uncertainty, tactical opportunity, and strategic habit. They therefore provide a strong empirical setting for examining personal informatics beyond domains where the tracked object is primarily bodily activity or biometric measurement.
Personal informatics refers to the collection and use of personally meaningful information for self-knowledge, reflection, and behavior change. Li et al. (2010) conceptualized personal informatics as a process through which people prepare to track, collect data, integrate those data, reflect on them, and act on what they learn. Later work complicated this stage-based view by showing that tracking is often lived, interrupted, socially shaped, and uneven rather than smoothly linear (Rooksby et al. 2014; Epstein et al. 2015, 2020). Sport-focused personal informatics research has similarly shown that tracking systems can support reflection and training decisions, but only when athletes and coaches can interpret data in relation to context, motivation, and practice (Rapp and Tirabeni 2018, 2020).
Chess extends this literature in an important direction. In many sport informatics settings, data are captured through wearables, sensors, or numerical performance dashboards. In chess, the primary data are symbolic and strategic: move sequences, positions, openings, engine evaluations, annotations, recalled lines of thought, and records of recurring mistakes. These data do not automatically explain themselves. An engine may identify a stronger move, but it does not necessarily explain why a human player failed to find it, how the position should have been understood, or how that insight should reshape future training. The central tension is therefore that chess generates unusually rich data, but rich data do not automatically generate learning. Chess self-analysis depends on interpretation as much as measurement.
This interpretive demand is especially salient among collegiate chess players. Student-athletes pursue development within the constraints of academic workload, tournament schedules, fatigue, limited coaching time, and uneven access to resources. Their practices are likely to include both systematic routines and practical compromises: recording some games but not others, keeping some archives while losing others, trusting engines in some positions while seeking human explanation in others. Such conditions make collegiate chess a valuable site for studying personal informatics as lived practice rather than as an idealized data pipeline.
Although chess produces unusually detailed records of cognitive performance, the availability of recorded moves, databases, and engine evaluations does not by itself explain how players convert those traces into self-knowledge or training action. Personal informatics research has shown how people collect and act on personal data, but it has largely emphasized bodily, behavioral, or sensor-generated tracking rather than symbolic records of decision-making. Collegiate chess provides a theoretically useful case because players must transform fragile records–notation sheets, memory reconstructions, PGNs, platform histories, annotations, and engine lines–into actionable interpretations while managing academic workload, tournament fatigue, uneven coaching access, and inconsistent archive practices. This study therefore examines how collegiate chess players construct, interpret, and use personal chess archives as symbolic performance informatics systems for perceived development.
The present study addresses five research questions:
- How do collegiate chess players capture, preserve, retrieve, and organize their own game records for later review?
- How do players transform recorded games into analyzable learning objects through replay, annotation, classification of errors, and comparison with external references?
- How do digital engines, databases, coaches, peers, and prior experience mediate players’ interpretation of their own games?
- What indicators do players use to judge whether self-analysis is contributing to perceived chess development, and how do they distinguish numerical indicators from experiential indicators?
- What breakdown and repair conditions interrupt, redirect, or sustain the movement from game capture to reflective judgment and training action?
This study makes three contributions. First, it develops symbolic performance informatics as an extension of personal informatics to domains where self-tracking centers on symbolic records of cognitive and strategic performance rather than bodily or sensor-generated metrics. Second, it identifies interpretive mediation as the mechanism through which engine outputs, databases, coaches, peers, and prior experience convert recorded games into usable training knowledge. Third, it positions personal chess archives as learning infrastructure, showing how capture, stabilization, retrieval, and annotation shape what players can infer from their own performances.
In this study, an analyzable learning object refers to a game record that has been made available for reflection through replay, annotation, classification, comparison, or contextual explanation.
Literature Review and Conceptual Framework
Personal Informatics and Lived Tracking
The Stage-Based Model of personal informatics remains a useful starting point for understanding how people turn personal data into action. Li et al. (2010) identify five interrelated stages: preparation, collection, integration, reflection, and action. The model is valuable because it emphasizes that data are not useful simply because they exist. Users must decide what to track, capture the relevant traces, organize them into usable form, interpret patterns, and translate insights into later behavior.
Subsequent research has shown that these stages rarely unfold in a perfectly ordered way. Rooksby et al. (2014) argue that personal tracking is better understood as lived informatics: tracking practices are adopted, adapted, interrupted, abandoned, and reconfigured within everyday life. Epstein et al. (2020) similarly show that barriers in one stage can shape or limit later stages. Incomplete collection can undermine reflection; weak integration can make action difficult; excessive or poorly interpreted data can reduce the value of tracking. These insights are crucial for the present study because collegiate chess players do not analyze games under laboratory conditions. Their recordkeeping is embedded in tournament pressure, academic schedules, fatigue, motivation, and access to tools or guidance.
Research on personal informatics in sport reinforces this point. Rapp and Tirabeni (2018) show that sport tracking systems can increase awareness and support training adjustment, but that athletes must still interpret data in relation to practice. Their later work also notes that self-tracking can affect comfort, motivation, attention, and lifestyle, while introducing risks of overload or overdependence on indicators (Rapp and Tirabeni 2020). These findings suggest that self-tracking is neither automatically beneficial nor purely technical. Its value depends on whether athletes can make sense of records and convert them into meaningful action.
Personal information management and personal archiving further clarify why chess records matter. Personal information management concerns how people keep, organize, retrieve, and reuse information for later goals (Jones 2007; Whittaker 2011). Personal archives are not neutral containers; people preserve materials because they anticipate future use, identity work, memory, or evidence (Kaye et al. 2006; Sinn et al. 2017). A personal chess archive can therefore be understood as a player-maintained collection of game records used for retrieval, review, comparison, preparation, or training decisions.
Reflective Learning and Cognitive Performance
Reflective learning and self-regulated learning provide a complementary foundation for understanding perceived chess development. Self-regulated learners set goals, monitor performance, evaluate outcomes, and adjust strategies in response to feedback (Winne 1997; Zimmerman 2000). Metacognitive monitoring allows learners to notice gaps between intention and performance, while reflective control allows them to change subsequent behavior (Efklides 2017; Higgins et al. 2021). In this sense, self-regulation and personal informatics share a common logic: both depend on a cycle linking record, reflection, and action.
Research on decision-making and reflective learning shows that learners can use reflection to evaluate alternatives and revise decisions (Gresch et al. 2015). However, reflection is not always constructive. Eikey et al. (2021) caution that self-tracking can become ruminative when users repeatedly inspect data without generating actionable understanding. This distinction is important for chess. Reviewing a lost game may produce insight, but it may also intensify frustration if the player cannot interpret why a position collapsed or what should change next.
Chess is a particularly strong domain for studying reflective learning because it creates stable records of performance. A game can be reconstructed move by move, allowing players to revisit critical decision points. Expertise research emphasizes that strong chess performance depends on pattern recognition, memory, flexible evaluation, and context-sensitive decision-making rather than calculation alone (Groot 1978; Chase and Simon 1973; Gobet and Simon 1996; Lafuente 2011). Game records therefore matter because they allow players to examine how decisions unfolded under particular constraints of time, complexity, and pressure.
Digital Tools and Chess Self-Analysis
Digital platforms have transformed chess learning. Lichess, Chess.com, ChessBase, Analyze This, CT-ART, Stockfish, and related systems allow players to store games, replay positions, compare alternatives, examine opening databases, solve puzzles, and receive engine feedback. These tools expand the scale and speed of self-analysis. They also create new forms of archive, allowing players to revisit years of games or share records with coaches and teammates.
Artificial intelligence is central to this shift. Gaessler and Piezunka (2023) argue that chess computers can act as artificial training partners by expanding access to high-level feedback and practice opportunities. This is especially relevant in resource-constrained settings where players may not always have a coach or strong sparring partner available. Yet engine output does not eliminate the need for human sense-making. A best move line may be objectively strong while remaining difficult for a player to understand, remember, or reproduce over the board. The literature therefore supports a view of digital chess tools as mediating technologies: they make analysis more available, but they do not by themselves complete the learning process.
This distinction also connects chess engines to broader human-AI explanation research. Explainable AI scholarship argues that users need outputs they can understand and appropriately trust, not only highly accurate recommendations (Gunning and Aha 2019). Work on algorithmic transparency similarly shows that the amount, complexity, and form of system information can shape trust and advice use rather than simply increasing reliance (Kizilcec 2016; Lehmann et al. 2022). In chess terms, a numerical evaluation or best-move sequence may identify a problem while leaving the human learner to reconstruct the strategic reason. This is why the present study treats engines as interpretive mediators rather than as self-sufficient teachers.
Recorded chess games also have value as information artifacts. Pickel and Rabern (2022) note that a properly recorded scorecard uniquely characterizes a chess game. Large-scale database studies show that game records preserve regularities, preferences, and memory effects across human play (Schaigorodsky et al. 2014, 2016). For individual players, this suggests that personal archives can reveal recurring tendencies and weaknesses. A player’s saved games are not inert history; they are structured traces that can be searched, replayed, compared, and transformed into training priorities.
Conceptual Framework
Figure 1 presents the Symbolic Performance Informatics Model of Collegiate Chess Self-Analysis. The model is adapted from the Stage-Based Model of personal informatics (Li et al. 2010), lived informatics (Rooksby et al. 2014; Epstein et al. 2015), personal information management (Jones 2007; Whittaker 2011), and self-regulated learning (Winne 1997; Zimmerman 2000). It defines a game record as any preserved representation of a chess game, including a notation sheet, memory reconstruction, PGN, platform history, study file, or annotated archive.
Symbolic performance informatics refers to self-tracking practices in which individuals capture, stabilize, annotate, interpret, and act on symbolic records of cognitive or strategic performance rather than sensor-generated bodily or behavioral metrics. This construct modifies personal informatics theory in two ways. First, it treats archive stabilization as analytically prior to reflection because symbolic records can be lost, incomplete, or unreplayable. Second, it places interpretive mediation between reflection and action because symbolic records require contextual explanation before they can guide future behavior. Traditional personal informatics emphasizes personal data collection and reflection; lived informatics emphasizes unevenness and interruption; personal information management emphasizes storage, retrieval, and reuse; and self-regulated learning emphasizes monitoring, judgment, and strategy adjustment. The present framework integrates these strands through symbolic records and the mediation required to make those records actionable.
The model reframes the informatics sequence for a symbolic cognitive-performance domain. Game capture refers to notation sheets, memory reconstruction, platform histories, and PGN exports. Archive stabilization refers to transferring those traces into replayable, retrievable, and durable personal records. Annotation and structuring refers to comments, symbolic marks, bookmarks, critical positions, and error categories. Interpretive mediation refers to the process by which players use engines, databases, coaches, peers, books, and prior experience to make sense of recorded positions. Self-regulatory judgment refers to how players identify weaknesses, set priorities, and evaluate confidence, readiness, and recurring patterns. Training action refers to opening revision, tactics practice, endgame study, time-management adjustment, and opponent preparation.
The framework also makes breakdown and repair analytically central. Fatigue, academic workload, incomplete notation, archive loss, inconsistent routines, and engine incomprehension can interrupt the movement from records to action. Repair refers to the practical work players use to restore that movement, such as encoding games immediately after a round, switching from paper to digital storage, asking a stronger player to explain engine lines, reviewing only critical positions when time is scarce, or delaying analysis until emotions have cooled. Perceived development is therefore treated as a player-identified outcome rather than a measured causal effect. It includes numerical indicators, such as ratings and fewer blunders, and experiential indicators, such as calmness, confidence, preparedness, and pattern recognition. Later review of perceived development feeds back into future capture, interpretation, judgment, and action.
The model is not proposed as a universal replacement for personal informatics frameworks. Rather, it specifies conditions under which personal informatics involves symbolic cognitive-performance records whose usefulness depends on preservation, replayability, interpretive mediation, and perceived transfer to future action.
Methodology
Research Design
The study employed a qualitative interpretive case-study design based on semi-structured interviews and artifact-elicitation prompts. This design fits the research problem because the study is concerned with meaning, practice, interpretation, and variation rather than statistical measurement or causal estimation. The research asks how players preserve records, how they make sense of those records, and how they report translating insights into future training. Such questions require attention to participant experience and contextual detail (Ng and White 2005).
Semi-structured interviews were used because they provide both consistency and flexibility. They allow the researcher to ask all participants about common domains such as notation, platforms, annotation, engine use, recurring errors, and development indicators, while also allowing follow-up questions that explore participant-specific routines or constraints (Adeoye-Olatunde and Olenik 2021; McIntosh and Morse 2015). This was important because collegiate chess players may share a general practice of reviewing games while differing substantially in tools, depth of analysis, reliance on guidance, and retention habits. Artifact-elicitation was used in a limited, privacy-preserving sense: participants were prompted to describe or walk through examples such as a notation sheet, online game archive, PGN file, Lichess study, Chess.com analysis page, annotated game, or difficult engine line when such examples were available and safe to discuss.
Participants and Data Source
The dataset consists of semi-structured interview transcripts from ten collegiate chess players affiliated with the University of the Philippines Diliman chess context. Participants were selected purposively because they had direct experience with competitive chess and with recording, storing, or analyzing their own games. Purposive sampling is appropriate in qualitative research when the goal is informational relevance rather than statistical representativeness (Patton 2015; Gill 2020; Koerber and McMichael 2008).
Inclusion criteria required participants to be current or recent collegiate chess players, to have competitive over-the-board or online tournament experience, to have recorded or reviewed at least some of their games, and to have used at least one self-analysis resource such as notation, PGN, Lichess, Chess.com, ChessBase, Stockfish, coach review, or peer analysis. Recruitment prioritized variation in competitive experience, preferred time control, engine use, coaching or peer support, archive sophistication, frequency of review, and role in team training. Ten participants were sufficient for this exploratory case study because the aim was focused, the sample was specific, participants had direct experience of the phenomenon, interviews generated detailed accounts of concrete practices, and the analysis sought thematic depth rather than population representation. This rationale follows the information-power view of qualitative sample adequacy (Malterud et al. 2016).
Participants were treated as key informants of practice rather than as units for demographic comparison. The transcript indicates that they had experience with tournament play, team training, online platforms, notation practices, engine analysis, opponent preparation, coach or teammate interaction, and rating-based performance monitoring. For confidentiality, the article uses anonymized participant labels rather than names. Direct quotations are presented selectively to illustrate analytic claims while avoiding unnecessary identifying detail.
Table 1 summarizes the anonymized participant variation used to interpret the findings. The table deliberately avoids exact ratings, usernames, named events, unique opening repertoires, and other details that could identify players within a small collegiate chess community.
| Participant | Competitive experience | Main format | Primary record type | Main analysis tools | Archive routine | Guidance source |
|---|---|---|---|---|---|---|
| P1 | Collegiate and earlier school tournaments | OTB standard; rapid by memory | Notation, memory reconstruction, digital copies | Analyze This, Lichess, Chess.com, Stockfish | Short-term review; some lost files | Solo; past coach; engine |
| P2 | Collegiate, online, and team play | Online plus OTB/team games | Notebook transferred to Lichess | Lichess, Chess.com, ChessBase | Long-term account-based retention | Brother; engine; solo |
| P3 | Collegiate and school tournaments | OTB and online study games | Paper notation and Lichess study | Lichess, CT-ART, ChessBase | Study files and paper copies | Software and external resources |
| P4 | Collegiate and national-level youth experience | OTB notation plus online review | Notation transferred to Lichess | Lichess, Chess.com, engine review | Long-term notation retention | Solo first, then engine |
| P5 | Collegiate tournament play | Recorded games and memorized positions | PGN and Lichess study | Lichess study and analysis mode | Long-term Lichess study | Solo with analysis mode |
| P6 | Collegiate and team training | OTB and digital replay | Analyze This and Lichess records | Analyze This, Lichess, engine comparison | Shift from paper to replayable digital records | Coach/teammate comments plus engine |
| P7 | Collegiate tournament play | OTB notation and digital study | Notation, Lichess study, comments | Lichess, ChessBase, engine notes | Study-based organization | Teammates and engine |
| P8 | Collegiate and formal tournament play | OTB and PGN-based review | Paper/pen and PGN files | Analyze This, PGN files, paper notes | Long-term PGN retention | Solo, coaches/teammates when stuck |
| P9 | Collegiate, online, and UAAP-related play | Online application of openings | Apps, PGNs, Lichess study | ChessBase, Lichess, Chess.com | Online archive and database routine | Master database and self-review |
| P10 | Collegiate and UAAP-related play | OTB notation and shared review | Notation sheets and private Lichess study | Lichess study, engine, coaches/teammates | Folder plus private Lichess study | Coaches, teammates, engine |
The matrix shows why the sample was analytically useful despite its small size: participants varied in capture method, archive durability, reliance on engines, and access to human guidance. These differences are used in the results to distinguish common patterns from boundary cases.
Interview Instrument and Data Collection
The interview guide was aligned with the research questions and the personal informatics framework. It included open-ended questions on chess background, preparation before matches, recording and organizing game records, annotation practices, analysis preferences, post-game review routines, time spent reviewing, motivations and barriers, record retention, recurring mistakes, development indicators, and the perceived role of tools in training decisions. This structure reflects recommended practice for semi-structured interview guides, which should connect the phenomenon, interview domains, and analytic focus while leaving room for participant language and examples (Kallio et al. 2016; O’Keeffe et al. 2016).
Interviews asked participants to describe concrete practices rather than abstract opinions alone. Participants explained how they recorded games during standard time controls, reconstructed rapid games from memory, transferred records into Lichess or other platforms, marked questionable moves, consulted engines or human guides, tracked ratings, and decided what to study next. When participants referred to artifacts, the analysis treated those accounts as elicited examples rather than as a complete documentary corpus. The study therefore makes claims about reported and elicited practices, not about the measured quality of participants’ full archives. The interviews were transcribed to create the textual dataset for analysis. The consent and data collection materials indicated voluntary participation, confidentiality protections, and the optional nature of sharing documents or information sources.
Interviews were conducted from March 18 to April 9, 2026, in face-to-face and online modes. The interview protocol and consent materials allowed participants to share scanned copies, screenshots, file exports, or links voluntarily, but the analysis did not require permanent collection of identifiable artifacts. Interviews used English and Filipino/Tagalog as participants preferred; excerpts originally given in Filipino/Tagalog were translated for reporting while retaining short source phrases when analytically useful. Transcript validation was used to support accuracy before thematic analysis.
Table 2 clarifies how artifact-elicitation entered the dataset. Because artifact sharing was voluntary and privacy-sensitive, the table reports artifact talk and elicited examples rather than a full archive inventory. Artifact examples were used to specify practices, confirm platform routines, and identify contradictions such as lost files or records that existed but were rarely revisited.
| Artifact type | Participants who discussed it | Used in analysis? | Notes |
|---|---|---|---|
| Notation sheet or notebook | 7 | Yes | Used to examine capture, transfer, and fragility of paper records. |
| Lichess study or platform archive | 8 | Yes | Used to examine stabilization, retrieval, replay, and private study routines. |
| Chess.com analysis or rating history | 4 | Yes | Used mainly for rating, puzzle, and progress-monitoring accounts. |
| PGN file or digital export | 4 | Yes | Used to examine portability and movement from event record to study file. |
| Annotated game or study note | 8 | Yes | Used to examine marks, comments, bookmarks, and critical-position labels. |
| Engine line or evaluation | 8 | Yes | Used to examine mediation, overreliance risk, and explanation gaps. |
Table 2 also shows a limit of the evidence. The study can analyze how participants described, displayed, or walked through record practices, but it cannot claim to have audited every participant’s complete archive.
Analytic Procedure and Trustworthiness
The analysis followed a reflexive thematic workflow moving from transcript immersion to coding, category development, theme construction, and interpretive writing. Thematic analysis is appropriate because it supports the identification of patterned meanings across qualitative data while allowing attention to context and variation (Nowell et al. 2017; Braun and Clarke 2022). The coding strategy was abductive: it combined inductive attention to participants’ own accounts with deductive sensitivity to personal informatics, lived informatics, personal information management, and self-regulated learning.
First-cycle coding focused on explicit practices, meanings, and constraints: notation, replay, PGN transfer, engine checking, coach consultation, archive loss, confidence, frustration, trust, readiness, fatigue, workload, missing records, and time pressure. Framework coding then mapped relevant codes to capture, stabilization, annotation, interpretive mediation, self-regulatory judgment, action, and breakdown. Theme construction emphasized mechanisms rather than only topics; for example, “engines produce evaluation, not understanding” was treated as stronger than a generic theme about digital tools. Negative cases were actively considered, including players who did not maintain durable archives, analyzed only selected losses, distrusted engine lines, or stopped after capture without sustained review.
Table 3 provides a compact audit trail from excerpt to theme. It is not a substitute for the full codebook, but it makes visible how raw accounts were converted into analytic claims.
| Excerpt cue | Initial code | Category | Final theme | Analytic interpretation |
|---|---|---|---|---|
| Immediately encoded remembered moves on a phone after rapid games | Immediate memory repair | Capture and repair | Game capture is fragile infrastructure | Capture depends on stabilizing fragile memory before the record disappears. |
| Added a question mark to a move even when the engine did not call it a blunder | Player-generated annotation | Annotation and structuring | Annotation converts records into learning objects | Annotation preserves the player’s own judgment alongside software evaluation. |
| Engine showed the sequence but did not explain why the move worked | Engine explanation gap | Interpretive mediation | Engines produce evaluation, not understanding | AI output becomes useful only after human or social explanation. |
| Records were kept in an app since 2019 but the copy was lost | Archive loss | Breakdown condition | Collegiate constraints interrupt self-analysis | A stored record is not useful if it cannot be retrieved when needed. |
| Used own analysis before checking engine evaluations | Independent judgment before engine | Self-regulatory judgment | Engines produce evaluation, not understanding | Some players actively protect metacognitive control from engine overreliance. |
The Stage-Based Model of personal informatics sensitized the analysis to preparation, collection, integration, reflection, and action, but it did not predetermine the themes. This distinction was important for confirmability. Trustworthiness was supported through close engagement with the transcript, procedural transparency, a visible codebook summary, analytic memoing, negative-case attention, and attention to variation rather than only recurrence (Cope 2013; Houghton et al. 2013; Thomas and Magilvy 2011). Dependability was strengthened by documenting the analytic movement from codes to categories to themes (Nowell et al. 2017; Janis 2022). The primary researcher worked within the same institutional context as the study and had prior familiarity with chess analysis tools, which created a risk of treating systematic self-analysis as more common or more desirable than participants experienced it. To reduce this risk, analytic memos tracked inconsistent review, archive loss, distrust of engines, laziness, fatigue, and non-use of tools as analytically significant cases rather than as deficiencies. Transferability was supported through contextual specificity: the findings are presented as grounded in one collegiate chess setting, not as statistically generalizable to all chess players.
| Research question | Primary coding domains | Linked results themes |
|---|---|---|
| RQ1: Capture, preservation, retrieval, and organization | Notation; memory reconstruction; platform histories; PGN transfer; archiving; retention; retrieval | Theme 1 |
| RQ2: Transformation into analyzable learning objects | Symbolic marks; text comments; bookmarks; move-by-move replay; critical positions; error categories | Themes 2 and 3 |
| RQ3: Interpretive mediation by tools and people | Engine checks; databases; coaches; teammates; siblings; books; master games; prior experience | Themes 1, 3, and 5 |
| RQ4: Numerical and experiential indicators of perceived development | Ratings; puzzle scores; fewer blunders; cleaner games; confidence; calmness; preparedness; pattern recognition | Theme 4 |
| RQ5: Breakdown and repair from capture to action | Fatigue; academic workload; schedule pressure; incomplete notation; lost archives; inconsistent routines; engine incomprehension; repair strategies | Themes 3 and 5 |
Table 4 makes explicit how each revised research question is carried into coding and theme construction. This map also shows that constraints were not treated as a minor limitation but as an analytic category that helps explain where the informatics process breaks down.
Ethics
Because the study involved human participants, ethical reporting required voluntary participation, informed consent, confidentiality, and fair representation. Qualitative interview ethics extend beyond formal consent because transcripts preserve participant experiences, judgments, routines, and vulnerabilities (Kara and Pickering 2017; Oye et al. 2015). The study therefore reports participant accounts anonymously, avoids unnecessary identifying detail, and treats the transcript as confidential research material. This is particularly important in a small collegiate chess context where team membership, tournament history, exact rating, distinctive opening repertoire, usernames, named coaches, or tournament histories could make participants recognizable (Damianakis and Woodford 2012). Any artifact examples were handled as optional elicitation prompts rather than as publishable screenshots unless they could be anonymized without exposing usernames, ratings, team roles, or other identifying details.
Results
Analysis of the interview transcript shows that collegiate chess players engage in post-game self-analysis through a structured but uneven system of symbolic performance informatics. They build and maintain records, replay positions, annotate mistakes, consult digital tools, compare alternatives, and report translating insights into later preparation or training. Five themes emerged from the analysis. Table 5 summarizes the themes, their central claims, the personal informatics processes they illuminate, and the kinds of evidence that support them.
| Theme | Core finding | Personal informatics process | Illustrative evidence |
|---|---|---|---|
| 1. Game capture is fragile infrastructure | Players capture, transfer, store, and retrieve games as personal performance records. | Capture, stabilization, retrieval | Notation sheets, memory-based reconstruction, Lichess studies, PGN files, notebooks, short- and long-term archives |
| 2. Annotation converts records into learning objects | Players make records useful by replaying moves, marking errors, writing comments, and identifying critical positions. | Annotation, structuring, reflection | Question marks, double question marks, text notes, bookmarks, move-by-move review, opening and endgame lessons |
| 3. Engines produce evaluation, not understanding | Platforms and engines expand access to evaluation, but players still need human interpretation and judgment. | Interpretive mediation, judgment | Lichess, Chess.com, Analyze This, ChessBase, CT-ART, Stockfish, coaches, siblings, teammates, master games |
| 4. Perceived development is inferred through mixed indicators | Progress is inferred through ratings and error reduction as well as confidence, calmness, and preparedness. | Self-regulatory judgment, monitoring | Blitz and puzzle ratings, fewer blunders, cleaner games, increased calmness, better pattern recognition |
| 5. Collegiate constraints interrupt self-analysis | The movement from record to action is limited by fatigue, workload, archive fragility, inconsistent routines, and interpretive difficulty. | Breakdown, repair, action | Motivation after losses, fatigue after many rounds, academic schedules, laziness, lost records, overreliance concerns |
Table 5 previews the results as a movement from capture to action, but it also highlights where that movement can stall. The table is therefore used as an analytic map for the prose that follows, not as a substitute for discussing the themes.
Theme 1: Game Capture Is Fragile Infrastructure
The first theme shows that collegiate chess self-analysis begins before interpretation. Players must first create and maintain an infrastructure through which games can become available for later review. Across the sample, capture ranged from paper notation and notebooks to Lichess studies, Chess.com histories, PGN files, ChessBase records, and private digital folders. These practices align with the capture and stabilization stages of the framework and with the preparation, collection, and integration stages of personal informatics (Li et al. 2010).
For standard over-the-board games, notation sheets functioned as the first layer of the personal archive. Participants described writing down moves during games, keeping hard copies from tournaments, and later transferring them into digital systems. In faster games, where complete notation was less feasible, memory became a temporary capture mechanism. P1 described the repair logic of immediate capture: “pagkatapos agad ng game, kukunin ko kaagad yung phone ko” and encode as many remembered moves as possible before the next game. This urgency reveals an important feature of chess data work: a game that is not stabilized quickly may disappear as an analyzable object.
Digital transfer was a second layer of infrastructure. Players moved games into Lichess, Chess.com, Analyze This, ChessBase, and related tools because these platforms made the games replayable, searchable, shareable, and easier to annotate. P2 described a common hybrid routine: “nagsusulat lang sa notebook tapos nililipat siya sa Lichess para may online record din.” P5 offered a more digital version: “I download my game’s PGN and analyze it to Lichess and make a Lichess study about it.” The movement from paper or memory into digital form corresponds to integration in the personal informatics process. A notation sheet can preserve a game, but a digital study can make that record interactive and durable.
Archive logic varied across participants. Some kept records “forever” or preserved physical folders and private studies across years. P4 reported keeping notation records for years, including older school records, because they made long-term progress and recurring mistakes observable. Others kept records for only three to six months or lost files and notebooks despite intending to save them. P1 described keeping games in an Analyze This file since 2019 but later losing the copy and not knowing where some hard-copy notation sheets were. This negative case matters because it shows that personal informatics in chess is not uniformly cumulative. Some players treat the archive as a long-term record of development and opponent preparation; others treat it as a short-term aid for extracting lessons while the game remains relevant. This supports lived informatics research by showing that tracking systems are shaped by convenience, motivation, and ordinary constraints rather than by a single ideal routine (Rooksby et al. 2014; Epstein et al. 2020). It also connects chess self-analysis to personal information management: keeping a record is meaningful only if the player can later find and reuse it (Jones 2007; Whittaker 2011).
The archive also served an anticipatory role. Several participants used stored games or databases to prepare for future opponents, review likely openings, or inspect recurring patterns in an opponent’s play. Thus, chess records were not only retrospective. They became resources for future decision-making. This finding strengthens the argument that self-analysis is a form of personal data work: players collect and organize their own performance traces so that those traces can later shape preparation, reflection, and action.
Theme 2: Annotation Converts Records Into Learning Objects
The second theme shows that records become useful only when players transform them through replay and annotation. Participants did not describe review as a passive glance at a final result. They described going move by move, identifying where the position changed, marking questionable moves, writing comments, bookmarking lines, and comparing alternatives. This is the point at which collected data become reflective resources.
Replay was central. P1 described reviewing a game by pressing through the moves sequentially, “next, next, next,” and then returning to the beginning once errors appeared. This practice reconstructs the game as a chain of decisions rather than a simple win or loss. It allows players to locate the moment where an opening choice became uncomfortable, a tactical opportunity was missed, or an endgame advantage began to disappear.
Annotation externalized judgment. Participants used question marks, double question marks, red marks, written comments, and bookmarks to distinguish blunders, disliked moves, unusual lines, and positions requiring future study. P2 emphasized that the player, not only the engine, could place a question mark on a move: “ako mismo yung naglalagay doon ng question mark kahit hindi siya blunder.” P4 similarly described using comments on critical moments and double question marks for blunders or missed moves. This kind of self-judgment is important because it preserves the player’s own interpretation before or alongside software evaluation. Annotation turns a momentary realization into a durable piece of personal information.
The content of reflection differed by phase of the game. Some participants focused strongly on openings, especially when preparing lines, avoiding unfamiliar variations, or comparing their choices with master games. Others identified middle-game planning, tactics, pawn structures, and conversion of advantages as recurring weaknesses. Endgame reflection often involved conceptual understanding, time pressure, and the difficulty of converting winning positions. P3 connected review to future study by saying that seeing the mistake becomes a guide for “ano yung aralin ko next.” Across these examples, reflection was attached to specific positions and decisions. Players did not merely say that they needed to “play better.” They identified parts of the game where future study should be directed.
This theme clarifies how personal informatics becomes actionable in chess. A record alone shows what happened. An annotated and replayed record shows what the player thinks mattered, what should be revisited, and what future action might follow. In this sense, annotation is not cosmetic. It is the mechanism through which a game becomes a personalized learning object.
Theme 3: Engines Produce Evaluation, Not Understanding
The third theme captures a central tension in the data. Digital tools clearly expanded what players could do with their records. Participants used Lichess, Chess.com, Analyze This, ChessBase, CT-ART, Stockfish, master databases, and puzzle tools to store games, locate errors, search sample games, review openings, test candidate moves, and monitor ratings. These tools reduced the burden of replay and evaluation, making post-game review faster and more accessible.
At the same time, participants repeatedly stated that tool output did not automatically become understanding. Engines could identify best moves or display a sequence, but several players noted that engines often did not explain why the move made sense in human terms. P2 contrasted engine output with help from a sibling: the engine “pinapakita lang yung tamang tira” but “hindi naman niya talaga na-explain kung bakit ganyan.” P5 made the same point in English: Lichess provides exact moves, “but it doesn’t discuss why, so I have to figure out on my own why it is accurate.” P10 valued master database games because they showed strategic planning across the middle game and endgame rather than only engine accuracy.
This tension complicates simple accounts of technology-assisted development. AI can function as an artificial training partner by providing scalable feedback (Gaessler and Piezunka 2023), but the interview data show that feedback must still be interpreted. Some participants preferred to analyze alone first so they could develop their own judgment before checking an engine. P4 described this as a deliberate sequence: reviewing without the engine first to test personal understanding, then checking objective evaluations. Others preferred guided analysis with teammates, coaches, or stronger family members because shared interpretation produced ideas that a machine line did not explain. These differences show that tool use is not a single practice. It is a negotiated relationship among algorithmic evaluation, personal judgment, and social interpretation.
The theme also highlights a risk of overreliance. Participants recognized that engines are powerful but “perfect” in a way that can distance analysis from over-the-board reality. P4 named a “tendency na masyado rin akong umaasa sa engine” and described balancing this by using personal analysis before engine evaluation. P10 similarly said that the engine shows the best line but is “masyadong perfect,” so master games can be more useful for seeing plans. A move that is best according to Stockfish may be difficult for a human to find, remember, or justify during a timed game. This finding connects with the emphasis on flexible decision-making in chess expertise (Lafuente 2011). Good analysis is not simply the discovery of a best move; it is the development of judgment that players believe can be used in future play.
Theme 4: Perceived Development Is Inferred Through Mixed Indicators
The fourth theme shows that players evaluate perceived development through a blended evidentiary system. Ratings mattered. Participants mentioned blitz ratings, online rating graphs, puzzle rush scores, FIDE or tournament ratings, and everyday progress as indicators of development. P2 summarized this most compactly as “Ratings. Ratings ng blitz and everyday progress.” P3 used Lichess rapid or blitz ratings as a target, while P9 compared rating before and after applying an opening over a month. These numerical signals provided visible evidence that practice might be working.
However, perceived development was not reducible to numbers. Participants also described progress as fewer blunders, cleaner games, better openings, improved tactical recognition, greater calmness, stronger confidence, and feeling more prepared after analysis. P5 linked analysis to emotional readiness: “It motivates me to be more prepared” and made the participant “calm during the game after analyzing.” P7 described confidence as a self-sensed indicator: when training increased, confidence rose; when training dropped, confidence also dropped. These experiential markers matter because they show that self-tracking in chess is both metric and interpretive.
This finding extends personal informatics research by showing that progress in cognitive sport is often recognized through felt readiness. A rating increase can be meaningful, but so can the experience of seeing a pattern faster, remaining calm under pressure, or understanding why a position should be played differently. P6 described comparing a newly encountered position with an older annotated record to see whether the earlier error had been avoided or repeated. P4 similarly treated cleaner computer-analysis reports, especially games with fewer blunders or mistakes, as evidence that decisions were becoming more controlled. Players treated records as evidence that learning might be occurring. In this way, self-analysis was perceived as both diagnostic and motivational, while remaining self-reported rather than objectively measured.
Theme 5: Collegiate Constraints Interrupt Self-Analysis
The fifth theme shows that self-analysis shaped players’ habits and dispositions, while also remaining vulnerable to interruption. Two subthemes clarify this pattern.
Self-Analysis Reshapes Learner Dispositions
Participants connected game analysis with patience, consistency, humility, critical thinking, and more disciplined study. P1 suggested that analyzing losses helped prevent arrogance because mistakes became visible: “hindi ka nagiging mayabang kasi nakikita mo yung mistakes mo.” P4 reported that regular tool-supported review made analysis more consistent and training more structured. P6 treated records as accountability, explaining that archived notes were “proof” of effort and made it easier to see what should be done next. These accounts do not prove behavioral transformation, but they show that players interpreted analysis as shaping how they approached learning.
These changes are important because they show that personal informatics may affect more than technical chess knowledge. Players do not only report learning an opening line or a better move. They may also become different kinds of learners: more reflective, more patient, more self-critical, and more deliberate. This supports the broader claim that personal informatics can shape behavior and identity, not simply provide data (Rapp and Tirabeni 2018).
Collegiate Conditions Interrupt and Redirect the Informatics Cycle
Yet the same participants also described constraints. Fatigue after many rounds, academic workload, laziness, schedule pressure, missing records, and difficulty interpreting engines all limited self-analysis. P1 explained that after many rapid rounds, continuous analysis could become exhausting: “nakakapagod na puro talo.” P7 linked missed analysis to academic scheduling, saying there were times when analysis stopped because “may need na akong gawin sa acads ko.” P5 described a boundary case where analysis became difficult without the exact copy of the position to review. These constraints show that self-analysis is not a frictionless cycle of optimization. It is an aspirational discipline negotiated within real student-athlete life.
Repair practices were therefore central. Players repaired fragile memory by encoding games immediately after rounds; repaired paper fragility by transferring notation into Lichess studies or PGN files; repaired engine incomprehension by asking siblings, coaches, teammates, or databases for explanation; and repaired limited time by reviewing only critical moments or returning to losses later. P6 described moving from paper openings to Lichess because the record could be replayed “anytime basta may device.” P10 described sharing Lichess studies with teammates or coaches because the convenience made the practice sustainable “in a long time.” Repair did not eliminate breakdown, but it helped redirect the process back toward reflection and action.
Taken together, the results show that collegiate chess players practice personal informatics through a complex relationship among records, tools, interpretation, and action. The value of a game record lies not in its mere existence but in the reflective system built around it. A notation sheet, Lichess study, or engine line becomes meaningful when it helps a player understand a recurring weakness, prepare a future response, or change a training habit.
Discussion
This study set out to examine how collegiate chess players use personal informatics practices in analyzing their games for perceived development. The findings show that post-game chess analysis is not a narrow technical routine. It is a broader system of symbolic performance informatics in which players capture games through notation or memory, transfer them into digital archives, annotate and replay them, interpret them with and against software, and report using them to guide later preparation, confidence, and training decisions. At the same time, these practices are uneven. They are shaped by fatigue, academic life, archive instability, interpretive difficulty, and variable reliance on digital tools.
The first contribution is symbolic performance informatics. Reviews of the field show that much of personal informatics research has centered on bodily activity, wellness indicators, and everyday habit tracking (Epstein et al. 2020). Chess offers a different case. Here, the data are move sequences, position histories, annotations, openings, engine evaluations, and records of error. These records are not biometric, but they still function as personal informatics because they are collected, maintained, interpreted, and used for self-knowledge and training regulation. The findings therefore support defining personal informatics by process rather than by data type alone, while adding that symbolic cognitive-performance records require stabilization and explanation before they become useful.
The second contribution is interpretive mediation between reflection and action. The findings strongly support the relevance of preparation, collection, integration, reflection, and action (Li et al. 2010), but the chess case shows that reflection does not move directly into action. A player must decide whether an engine line is understandable, whether it fits the player’s repertoire, whether the error was tactical, strategic, emotional, or time-related, and whether a lesson can be practiced. This mediation happens through engines, databases, coaches, peers, books, and prior experience. It is the theoretical hinge between reflection and action in Figure 1 and helps explain why stronger evaluation is not the same as usable learning.
The theoretical implication is that, in symbolic performance informatics, the central problem is not only whether users collect data, but whether they can stabilize and mediate symbolic traces sufficiently for future action.
The third contribution is personal archives as learning infrastructure. Chess records are not passive storage. They are personal information systems in use. A raw game record shows what happened, but an annotated and replayed record shows what the player thinks mattered. This distinction has implications for Library and Information Science because it shows that even outside formal institutional contexts, people build personal archives and retrieval systems to support learning, decision-making, and expertise (Jones 2007; Kaye et al. 2006; Sinn et al. 2017). The quality of capture, stabilization, retrieval, and annotation shapes what players can infer from their own records.
The findings also complicate how development is recognized. Participants used ratings and error reduction, but they also trusted felt indicators such as calmness, preparedness, confidence, and pattern recognition. Players treated these mixed indicators as signs of development when numerical evidence and lived readiness began to converge. This remains a self-reported judgment, not an objective demonstration of performance gain.
These findings have practical implications grounded in the results. Players should use immediate post-game capture because memory reconstruction was fragile and games could disappear if not stabilized quickly. They should conduct independent replay before engine use because several participants valued testing their own judgment before checking evaluations. Coaches should teach human explanation of engine lines because participants found engines useful but insufficiently explanatory.
University programs should recognize post-game analysis as student-athlete learning labor because fatigue and academic workload interrupted review routines. Tool designers should build features that connect recurring errors across games, explain engine lines in human terms, and translate analysis into future study priorities because participants struggled to move from evaluation to actionable understanding.
The study is limited by its qualitative and context-bound design. It is based on interviews with collegiate players in one institutional setting and does not measure the causal effect of self-analysis on rating gain. It also relies primarily on participant accounts and elicited examples rather than direct longitudinal analysis of participants’ full game archives. Because artifacts were not systematically collected, retained, and independently analyzed across all participants, the study treats artifact-elicitation as a way to deepen interview accounts rather than as a separate documentary analysis. Future research should compare self-analysis practices across novice, collegiate, and elite players; examine actual PGN files, notation sheets, annotations, and study folders alongside interviews; and investigate whether particular archive and annotation routines correspond to long-term performance development.
Conclusion
Collegiate chess players use game records as personal informatics resources for perceived development. They capture games, preserve them in personal archives, annotate and replay critical positions, negotiate the meaning of engine output, monitor progress through both numbers and felt readiness, and report adjusting training in response to what they learn. These practices show that personal informatics is not confined to wearable devices or bodily metrics. It also exists wherever people systematically preserve traces of their own performance and use those traces to become more reflective, strategic, and disciplined learners.
The study’s central claim is that collegiate chess self-analysis is a form of symbolic performance informatics. Game records become useful only when they are stabilized, annotated, interpreted, and translated into training action under real institutional and personal constraints. Its success depends on more than access to tools. It depends on the player’s ability to create durable records, interpret them meaningfully, connect insights across time, and act on those insights under the real constraints of collegiate life.
Appendix: Codebook Summary
The codebook summary in Table 6 reports the main analytic categories used to move from first-cycle codes to themes. It is intentionally summarized for article format; the full working codebook contains more granular codes, inclusion criteria, exclusions, and exemplar quotations.
| Category | Example codes | Analytic purpose | Linked theme(s) |
|---|---|---|---|
| Pre-match preparation and anticipatory tracking | Opening preparation; opponent-specific preparation; use of external learning resources | Identified how players prepare games and opponents as future data sources. | Theme 1 |
| Capture, storage, and retrieval of game data | Notation-based capture; memory-based reconstruction; immediate post-game encoding; digital transfer; platform-based archiving; hybrid storage; retention; archive loss | Captured the infrastructure that makes games available for later reflection. | Theme 1 |
| Reflective review and annotation practices | Symbolic annotation; written comments; reflective replay; self-judgment of moves; opening review; middle-game weakness; endgame conversion failure | Explained how records become actionable through added judgment and structure. | Theme 2 |
| Interpretive mediation and guidance | Engine-assisted evaluation; engine skepticism; human-guided interpretation; solo analysis preference; guided analysis preference | Tracked how players negotiate machine, personal, and social sources of meaning. | Theme 3 |
| Perceived development and mixed indicators | Progress through rating; progress through error reduction; progress through confidence and calmness; pattern recognition | Showed how players decide whether self-analysis is contributing to perceived development. | Theme 4 |
| Motivations, barriers, and behavioral consequences | Motivation for development; competitive identity; fatigue; schedule pressure; laziness; interpretive difficulty; consistency; patience; humility; overreliance risk | Located the conditions that sustain, reshape, or interrupt self-tracking. | Theme 5 |
References
Adeoye-Olatunde, Omolola A., and Nicole L. Olenik. 2021. “Research and Scholarly Methods: Semi-Structured Interviews.” JACCP: Journal of the American College of Clinical Pharmacy 4 (10): 1358–67. https://doi.org/10.1002/jac5.1441.
Braun, Virginia, and Victoria Clarke. 2022. “Conceptual and Design Thinking for Thematic Analysis.” Qualitative Psychology 9 (1): 3–26. https://doi.org/10.1037/qup0000196.
Chase, William G., and Herbert A. Simon. 1973. “Perception in Chess.” Cognitive Psychology 4 (1): 55–81. https://doi.org/10.1016/0010-0285(73)90004-2.
Cope, Diane G. 2013. “Methods and Meanings: Credibility and Trustworthiness of Qualitative Research.” Oncology Nursing Forum 41 (1): 89–91. https://doi.org/10.1188/14.ONF.89-91.
Damianakis, Thecla, and Michael R. Woodford. 2012. “Qualitative Research with Small Connected Communities.” Qualitative Health Research 22 (5): 708–18. https://doi.org/10.1177/1049732311431444.
Efklides, Anastasia. 2017. “Affect, Epistemic Emotions, Metacognition, and Self-Regulated Learning.” Teachers College Record 119 (13): 1–22. https://doi.org/10.1177/016146811711901302.
Eikey, Elizabeth V., Clara Caldeira, Mayara Costa Figueiredo, Yunan Chen, and Daniel A. Epstein. 2021. “Beyond Self-Reflection: Introducing the Concept of Rumination in Personal Informatics.” Personal and Ubiquitous Computing 25 (3): 601–16. https://doi.org/10.1007/s00779-021-01573-w.
Epstein, Daniel A., Clara Caldeira, Mayara Costa Figueiredo, et al. 2020. “Mapping and Taking Stock of the Personal Informatics Literature.” Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 4 (4): 1–38. https://doi.org/10.1145/3432231.
Epstein, Daniel A., An Ping, James Fogarty, and Sean A. Munson. 2015. “A Lived Informatics Model of Personal Informatics.” Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing, 731–42. https://doi.org/10.1145/2750858.2804250.
Gaessler, Fabian, and Henning Piezunka. 2023. “Training with AI: Evidence from Chess Computers.” Strategic Management Journal 44 (11): 2724–50. https://doi.org/10.1002/smj.3512.
Gill, S. L. 2020. “Qualitative Sampling Methods.” Journal of Human Lactation 36 (4): 579–81. https://doi.org/10.1177/0890334420949218.
Gobet, Fernand, and Herbert A. Simon. 1996. “Templates in Chess Memory: A Mechanism for Recalling Several Boards.” Cognitive Psychology 31 (1): 1–40. https://doi.org/10.1006/cogp.1996.0011.
Gresch, Helge, Marcus Hasselhorn, and Susanne Boegeholz. 2015. “Enhancing Decision-Making in STSE Education by Inducing Reflection and Self-Regulated Learning.” Research in Science Education 47 (1): 95–118. https://doi.org/10.1007/s11165-015-9491-9.
Groot, Adriaan D. de. 1978. Thought and Choice in Chess. Mouton.
Gunning, David, and David Aha. 2019. “DARPA’s Explainable Artificial Intelligence (XAI) Program.” AI Magazine 40 (2): 44–58. https://doi.org/10.1609/aimag.v40i2.2850.
Higgins, Nick, Sarah Frankland, and Joseph A. Rathner. 2021. “Self-Regulated Learning in Undergraduate Science.” International Journal of Innovation in Science and Mathematics Education 29 (1): 58–76. https://doi.org/10.30722/ijisme.29.01.005.
Houghton, Catherine, Dympna Casey, David Shaw, and Kathy Murphy. 2013. “Rigour in Qualitative Case-Study Research.” Nurse Researcher 20 (4): 12–17. https://doi.org/10.7748/nr2013.03.20.4.12.e326.
Janis, Irving. 2022. “Strategies for Establishing Dependability Between Two Qualitative Intrinsic Case Studies: A Reflexive Thematic Analysis.” Field Methods 34 (3): 240–55. https://doi.org/10.1177/1525822X211069636.
Jones, William. 2007. “Personal Information Management.” Annual Review of Information Science and Technology 41 (1): 453–504. https://doi.org/10.1002/aris.2007.1440410117.
Kallio, Hanna, Anna-Maija Pietila, Martin Johnson, and Mari Kangasniemi. 2016. “Systematic Methodological Review: Developing a Framework for a Qualitative Semi-Structured Interview Guide.” Journal of Advanced Nursing 72 (12): 2954–65. https://doi.org/10.1111/jan.13031.
Kara, Helen, and Lucy Pickering. 2017. “New Directions in Qualitative Research Ethics.” International Journal of Social Research Methodology 20 (3): 239–41. https://doi.org/10.1080/13645579.2017.1287869.
Kaye, Joseph Jofish, Janet Vertesi, Shari Avery, et al. 2006. “To Have and to Hold: Exploring the Personal Archive.” Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 275–84. https://doi.org/10.1145/1124772.1124814.
Kizilcec, Rene F. 2016. “How Much Information? Effects of Transparency on Trust in an Algorithmic Interface.” Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems, 2390–95. https://doi.org/10.1145/2858036.2858402.
Koerber, Amy, and Lonie McMichael. 2008. “Qualitative Sampling Methods.” Journal of Business and Technical Communication 22 (4): 454–73. https://doi.org/10.1177/1050651908320362.
Lafuente, Victor de. 2011. “Flexible Decisions and Chess Expertise.” Frontiers in Neuroscience 5. https://doi.org/10.3389/fnins.2011.00004.
Lehmann, Cedric Alexander, Christiane B. Haubitz, Andreas Fuegener, and Ulrich W. Thonemann. 2022. “The Risk of Algorithm Transparency: How Algorithm Complexity Drives the Effects on the Use of Advice.” Production and Operations Management 31 (9): 3419–34. https://doi.org/10.1111/poms.13770.
Li, Ian, Anind Dey, and Jodi Forlizzi. 2010. “A Stage-Based Model of Personal Informatics Systems.” Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 557–66. https://doi.org/10.1145/1753326.1753409.
Malterud, Kirsti, Volkert Dirk Siersma, and Ann Dorrit Guassora. 2016. “Sample Size in Qualitative Interview Studies: Guided by Information Power.” Qualitative Health Research 26 (13): 1753–60. https://doi.org/10.1177/1049732315617444.
McIntosh, Michele J., and Janice M. Morse. 2015. “Situating and Constructing Diversity in Semi-Structured Interviews.” Global Qualitative Nursing Research 2. https://doi.org/10.1177/2333393615597674.
Ng, C. K. C., and Peter White. 2005. “Qualitative Research Design and Approaches in Radiography.” Radiography 11 (3): 217–25. https://doi.org/10.1016/j.radi.2005.03.006.
Nowell, Lorelli S., Jill M. Norris, Deborah E. White, and Nancy J. Moules. 2017. “Thematic Analysis: Striving to Meet the Trustworthiness Criteria.” International Journal of Qualitative Methods 16 (1). https://doi.org/10.1177/1609406917733847.
O’Keeffe, Jimmy, Wouter Buytaert, and Ana Mijic. 2016. “The Use of Semi-Structured Interviews for the Characterisation of Farmer Irrigation Practices.” Hydrology and Earth System Sciences 20 (5): 1911–24. https://doi.org/10.5194/hess-20-1911-2016.
Oye, Christine, Niels O. Sorensen, and Stinne Glasdam. 2015. “Qualitative Research Ethics on the Spot.” Nursing Ethics 23 (4): 455–64. https://doi.org/10.1177/0969733014567023.
Patton, Michael Quinn. 2015. Qualitative Research and Evaluation Methods. 4th ed. SAGE.
Pickel, Bryan, and Brian Rabern. 2022. “Scorekeeping in a Chess Game.” Semantics and Pragmatics 15 (12): 1–21. https://doi.org/10.3765/sp.15.12.
Rapp, Amon, and Lia Tirabeni. 2018. “Personal Informatics for Sport: Meaning, Body, and Social Relations in Amateur and Elite Athletes.” ACM Transactions on Computer-Human Interaction 25 (3): 1–30. https://doi.org/10.1145/3196829.
Rapp, Amon, and Lia Tirabeni. 2020. “Self-Tracking While Doing Sport: Comfort, Motivation, Attention and Lifestyle of Athletes Using Personal Informatics Tools.” International Journal of Human-Computer Studies 140: 102434. https://doi.org/10.1016/j.ijhcs.2020.102434.
Rooksby, John, Mattias Rost, Alistair Morrison, and Matthew C. Chalmers. 2014. “Personal Tracking as Lived Informatics.” Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 1163–72. https://doi.org/10.1145/2556288.2557039.
Schaigorodsky, Ana L., Juan I. Perotti, and Orlando V. Billoni. 2014. “Memory and Long-Range Correlations in Chess Games.” Physica A: Statistical Mechanics and Its Applications 394: 304–11. https://doi.org/10.1016/j.physa.2013.09.035.
Schaigorodsky, Ana L., Juan I. Perotti, and Orlando V. Billoni. 2016. “A Study of Memory Effects in a Chess Database.” PLOS ONE 11 (12): e0168213. https://doi.org/10.1371/journal.pone.0168213.
Sinn, Donghee, Soojung Kim, and Sue Yeon Syn. 2017. “Personal Digital Archiving: Influencing Factors and Challenges to Practices.” Library Hi Tech 35 (2): 222–39. https://doi.org/10.1108/LHT-09-2016-0103.
Thomas, Eileen, and Joan K. Magilvy. 2011. “Qualitative Rigor or Research Validity in Qualitative Research.” Journal for Specialists in Pediatric Nursing 16 (2): 151–55. https://doi.org/10.1111/j.1744-6155.2011.00283.x.
Whittaker, Steve. 2011. “Personal Information Management: From Information Consumption to Curation.” Annual Review of Information Science and Technology 45 (1): 1–62. https://doi.org/10.1002/aris.2011.1440450108.
Winne, Philip H. 1997. “Experimenting to Bootstrap Self-Regulated Learning.” Journal of Educational Psychology 89 (3): 397–410. https://doi.org/10.1037/0022-0663.89.3.397.
Zimmerman, Barry J. 2000. “Attaining Self-Regulation: A Social Cognitive Perspective.” In Handbook of Self-Regulation, edited by Monique Boekaerts, Paul R. Pintrich, and Moshe Zeidner. Academic Press. https://doi.org/10.1016/B978-012109890-2/50031-7.