\documentclass[11pt]{article}
\usepackage[a4paper,margin=2cm,headheight=15pt,headsep=10pt]{geometry}
\usepackage{fontspec}
\setmainfont{TeX Gyre Termes}
\setsansfont{TeX Gyre Heros}
\newfontfamily\hf{TeX Gyre Heros}
\usepackage[table,dvipsnames]{xcolor}
\definecolor{navy}{HTML}{2F5496}
\definecolor{navyd}{HTML}{1F3864}
\definecolor{navylt}{HTML}{EEF2F9}
\definecolor{rulegray}{HTML}{9AA6B8}
\usepackage{amssymb}
\usepackage{booktabs,array,tabularx,longtable,multirow,colortbl}
\usepackage{ragged2e}
\usepackage{enumitem}
\setlist{nosep,leftmargin=1.4em}
\usepackage{caption}
\captionsetup{font=small,labelfont={bf,sf},labelsep=period,skip=5pt}
\usepackage{titlesec}
\titleformat{\section}{\Large\bfseries\sffamily\color{navy}}{\makebox[1.4em][l]{\thesection}}{0em}{}
\titlespacing*{\section}{0pt}{16pt}{6pt}
\titleformat{\subsection}{\large\bfseries\sffamily\color{navyd}}{\thesubsection}{0.5em}{}
\titlespacing*{\subsection}{0pt}{10pt}{3pt}
\usepackage{pdflscape}
\usepackage{fancyhdr}
\usepackage[hidelinks]{hyperref}
\pagestyle{fancy}\fancyhf{}
\renewcommand{\headrulewidth}{0.4pt}
\renewcommand{\footrulewidth}{0pt}
\fancyhead[L]{\footnotesize\sffamily\color{navy}Shared-access denial \& the individual-attribution gap}
\fancyhead[R]{\footnotesize\sffamily\color{navy}Instruments \& dataset}
\fancyfoot[C]{\footnotesize\sffamily\thepage}
\renewcommand{\arraystretch}{1.28}
\setlength{\arrayrulewidth}{0.4pt}
\arrayrulecolor{rulegray}
\newcolumntype{L}[1]{>{\RaggedRight\arraybackslash}p{#1}}
\newcolumntype{C}[1]{>{\Centering\arraybackslash}p{#1}}
\newcommand{\hc}[1]{{\hf\bfseries\color{white}#1}}
\newcommand{\hrow}{\rowcolor{navy}}
\newcommand{\tnote}[1]{\par\vspace{2pt}{\small\itshape #1}}
\hypersetup{pdftitle={Research Instruments and Analytic Dataset},pdfauthor={Alshammari and Alfraidi}}
\begin{document}

\thispagestyle{empty}
\begin{center}
{\hf\bfseries\color{navy}\LARGE Research Instruments and Analytic Dataset}\\[6pt]
{\hf\large\color{navyd} Supplementary methodological materials}\\[14pt]
{\itshape\large Shared-access denial and the individual-attribution gap\\ in Saudi cybercrime judgments}\\[12pt]
{\normalsize Bunder Sebail Alshammari$^{1,*}$ \quad and \quad Khaled Nasser Alfraidi$^{1}$}\\[4pt]
{\small\itshape $^{1}$Department of Arabic Language, College of Arts and Letters, University of Ha'il, Ha'il, Saudi Arabia}\\[2pt]
{\small $^{*}$Corresponding author: ban.alshammari@uoh.edu.sa}
\end{center}
\vspace{4pt}
\noindent\textcolor{navy}{\rule{\linewidth}{1pt}}
\vspace{4pt}

\noindent This document consolidates the study's three research instruments and its complete analytic dataset in a single reference. The corpus comprises twelve authenticated judgments (Cases~A--L) from Eastern Province judicial circuits, 1432--1435~AH (2011--2014), adjudicated under the Anti-Cyber Crime Law. The design is qualitative and information-rich: reported frequencies are descriptive characterisations of a small purposive corpus, not inferential population estimates. All identifiers are anonymised; no party-identifying details are stored.

\vspace{6pt}
\noindent\textbf{Contents.} The materials are organised as: the case-selection rubric (Instrument~1); the forensic-pragmatic coding scheme (Instrument~2); the judicial-uptake scale (Instrument~3); a variable dictionary; the twelve-case master dataset; a worked application of the Multidimensional Discourse Integration Model (MDIM); the segment-level inter-coder reliability record and Cohen's~$\kappa$ computation; and descriptive strategy, uptake, and co-occurrence distributions.

\vspace{6pt}
\begin{center}\small
\begin{tabular}{@{}L{5.2cm}L{10.4cm}@{}}
\hrow \hc{Machine-readable file} & \hc{Contents} \\
\texttt{\ldots\_instrument\_and\_data.xlsx} & Eleven-sheet workbook (instruments, dictionary, data, statistics). \\
\rowcolor{navylt}\texttt{cybercrime\_corpus\_master.sav} & SPSS: 12 cases $\times$ 11 coded variables (value-labelled). \\
\texttt{intercoder\_reliability.sav} & SPSS: 50 coded segments $\times$ two coders. \\
\end{tabular}
\end{center}
\tnote{The uploaded \texttt{intercoder\_reliability\_1.sav} is byte-for-byte identical to \texttt{intercoder\_reliability.sav} and is treated here as a single file.}
\clearpage

\section{Instrument 1 — Case selection and inclusion rubric}
Screening reduced the forty-seven cybercrime judgments in the source volume to nineteen cases with explicit defendant responses to digital evidence, and then to the twelve analytic cases that satisfied every criterion below.
\vspace{4pt}
\begin{center}\small\begin{tabular}{@{}L{3.4cm}L{10.6cm}C{1.7cm}@{}}
\hrow \hc{Criterion} & \hc{Requirement} & \hc{Verification} \\
Legal basis & Adjudicated under the Anti-Cyber Crime Law (Royal Decree M/17). & Yes\,/\,No \\
\rowcolor{navylt}
Technical trace & Evidence involves an IP address, MAC address, or account log. & Yes\,/\,No \\
Defence type & Defendant explicitly invokes shared infrastructure (Wi-Fi, household PC, shared account). & Yes\,/\,No \\
\rowcolor{navylt}
Discursive data & Judgment includes verbatim or detailed paraphrased speech of the defendant. & Yes\,/\,No \\
Judicial logic & Judge explicitly discusses doubt (shubha), intent (niyya), or evidence weight. & Yes\,/\,No \\
\end{tabular}\end{center}
\begin{landscape}
\section{Instrument 2 — Forensic-pragmatic coding scheme}
Five defence strategies structure the coding of every extracted segment. Arabic linguistic triggers are illustrative diagnostic cues, not exhaustive lists.
\vspace{4pt}
\begin{center}\footnotesize\begin{tabular}{@{}C{0.7cm}L{1.7cm}L{3.0cm}L{4.5cm}L{5.0cm}L{4.6cm}L{2.7cm}@{}}
\hrow \hc{\#} & \hc{Code} & \hc{Strategy} & \hc{Operational definition} & \hc{Arabic triggers (examples)} & \hc{Decision rule} & \hc{Jurisprudential hinge} \\
1 & \textbf{CAT-DEN} & Categorical denial & Complete rejection of the act and of authorship. & \emph{lam af'al (I did not do); ma li 'alaqa (I have no connection).} & Dominant move denies the act itself rather than the trace-to-actor inference. & Sufficiency and decisiveness of proof \\
\rowcolor{navylt}
2 & \textbf{SAB-TRANS} & Shared-access blame transfer & Concedes the trace location but names other users with access. & \emph{al-shabaka mushtaraka (shared network); ahl al-bayt (family members).} & Defendant admits the environmental link yet foregrounds alternative users. & shubha (legal doubt) \\
3 & \textbf{EVID-DISP} & Evidential displacement & Attacks the trace-to-actor inference itself. & \emph{al-taqrir la yuthbit (the report does not prove [me]); mujarrad ittisal (merely a connection).} & Central move contests the granularity or reach of the technical report. & Threshold of decisive proof \\
\rowcolor{navylt}
4 & \textbf{MOR-REL} & Moral-religious credibility work & Uses piety and character to bolster truth-claims. & \emph{Wallahi (by God); akhsha Allah (I fear God); laysa min khuluqi (not my character).} & Moral-religious positioning is the salient support for the denial. & niyya and credibility \\
5 & \textbf{ALT-CAUS} & Alternative-causation narrative & Proposes a specific external event such as hacking or framing. & \emph{ikhtiraq (hacking); talfiq (framing / fabrication).} & Defence advances a positive alternative causal story. & Alternative causation, shubha, intent \\
\end{tabular}\end{center}
\vspace{2pt}\small\textbf{Assignment procedure.} One \emph{primary} strategy is assigned per case by three ordered rules: (1)~frequency within the defendant's response; (2)~explicitness of judicial evaluation; (3)~argumentative centrality to the final ruling. Secondary strategies are recorded when present. \textbf{Reliability.} Two coders independently coded 50 extracted segments; Cohen's~$\kappa = 0.92$ (almost perfect, Landis \& Koch 1977); disagreements were adjudicated by a third reviewer.
\end{landscape}
\section{Instrument 3 — Judicial uptake evaluation scale}
Each judgment receives one uptake level recording how the court treated the shared-access denial.
\vspace{4pt}
\begin{center}\small\begin{tabular}{@{}C{1.3cm}L{4.3cm}L{9.9cm}@{}}
\hrow \hc{Level} & \hc{Judicial uptake category} & \hc{Operational definition} \\
0 & Rejection & Judge treats the technical trace as conclusive and dismisses the denial as a bare claim. \\
\rowcolor{navylt}
1 & Procedural recognition & Judge acknowledges shared access but finds it outweighed by other evidence. \\
2 & Mitigation / doubt & Judge admits shubha regarding the specific actor and reduces the ta'zir sentence. \\
\rowcolor{navylt}
3 & Non-conviction & Judge finds the evidence insufficient for individual attribution; dismissal or acquittal. \\
\end{tabular}\end{center}
\section{Variable dictionary}
\subsection*{\texttt{cybercrime\_corpus\_master.sav}}
\begin{center}\small\begin{tabular}{@{}L{3.0cm}L{4.6cm}L{7.9cm}@{}}
\hrow \hc{Variable} & \hc{Label} & \hc{Value codes} \\
\texttt{case\_num} & Case number (1--12) & 1--12 \\
\rowcolor{navylt}
\texttt{case\_id} & Case identifier (A--L) & A--L (string) \\
\texttt{year\_hijri} & Year of judgment (Hijri) & 1432--1435 \\
\rowcolor{navylt}
\texttt{offence\_code} & Dominant offence context & 1=Defamation / posting attribution; 2=Privacy / reputational posting; 3=Online threat / abusive message; 4=Blackmail / coercive communication; 5=Unauthorised access allegation; 6=Technical-report attribution dispute; 7=Account-use attribution dispute; 8=Hacking / framing scenario \\
\texttt{trace\_code} & Technical trace type & 1=IP Log (Household); 2=PC Forensic Image; 3=Mobile Device Log; 4=Account Metadata; 5=IP Log (Open Wi-Fi); 6=Social Media Log; 7=Router Admin Log; 8=Forensic Report; 9=Login History; 10=Device Ownership; 11=Account Logs; 12=Remote Access Log \\
\rowcolor{navylt}
\texttt{feature\_code} & Shared-access feature & 1=Household router / network; 2=Shared family computer; 3=Communal mobile / device access; 4=Shared account / device access; 5=Open or weakly secured Wi-Fi; 6=IP log without user-level corroboration; 7=Device logs without individual identification; 8=Unknown third-party network exploitation \\
\texttt{shared\_access} & Shared digital infrastructure involved & 0=No; 1=Yes \\
\rowcolor{navylt}
\texttt{primary\_code} & Primary defence strategy & 1=CAT-DEN (Categorical denial); 2=SAB-TRANS (Shared-access blame transfer); 3=EVID-DISP (Evidential displacement); 4=MOR-REL (Moral-religious credibility work); 5=ALT-CAUS (Alternative-causation narrative) \\
\texttt{hinge\_code} & Islamic legal hinge invoked & 1=Shubha (legal doubt); 2=Bayyinah (proof sufficiency); 3=Istishab (presumption); 4=Niyya (intent); 5=Ta'zir discretion; 6=Yaqin (certainty); 7=Tazkiyah (credibility) \\
\rowcolor{navylt}
\texttt{uptake\_level} & Judicial uptake level (0--3) & 0=Rejection; 1=Procedural recognition; 2=Mitigation / doubt; 3=Non-conviction \\
\texttt{conviction} & Conviction recorded & 0=No; 1=Yes \\
\end{tabular}\end{center}
\subsection*{\texttt{intercoder\_reliability.sav}}
\begin{center}\small\begin{tabular}{@{}L{3.0cm}L{4.6cm}L{7.9cm}@{}}
\hrow \hc{Variable} & \hc{Label} & \hc{Value codes} \\
\texttt{segment\_id} & Coded defence segment ID & S01--S50 (string) \\
\rowcolor{navylt}
\texttt{case\_id} & Case identifier (A--L) & A--L (string) \\
\texttt{coderA\_code} & Coder A (adjudicated) strategy & 1=CAT-DEN (Categorical denial); 2=SAB-TRANS (Shared-access blame transfer); 3=EVID-DISP (Evidential displacement); 4=MOR-REL (Moral-religious credibility work); 5=ALT-CAUS (Alternative-causation narrative) \\
\rowcolor{navylt}
\texttt{coderB\_code} & Coder B strategy & 1=CAT-DEN (Categorical denial); 2=SAB-TRANS (Shared-access blame transfer); 3=EVID-DISP (Evidential displacement); 4=MOR-REL (Moral-religious credibility work); 5=ALT-CAUS (Alternative-causation narrative) \\
\texttt{agreement} & Coders agree (1 = yes) & 0=No; 1=Yes \\
\end{tabular}\end{center}
\begin{landscape}
\section{Master analytic dataset — twelve judgments (Cases A--L)}
\vspace{2pt}
\begin{center}\scriptsize\begin{tabular}{@{}C{0.5cm}C{0.9cm}C{0.9cm}L{2.9cm}L{2.4cm}L{2.9cm}C{1.5cm}L{2.3cm}C{1.6cm}C{0.9cm}L{3.2cm}@{}}
\hrow \hc{Case} & \hc{Yr AH} & \hc{Yr CE} & \hc{Offence context} & \hc{Trace type} & \hc{Shared-access feature} & \hc{Primary} & \hc{Islamic hinge} & \hc{Uptake} & \hc{Conv.} & \hc{Judicial outcome (detail)} \\
A & 1433 & 2012--2013 & Defamation / posting attribution & IP Log (Household) & Household router / network & SAB-TRANS & Shubha (legal doubt) & 2 — Mitigation / doubt & No & Mitigation / doubt-constrained \\
\rowcolor{navylt}
B & 1432 & 2011--2012 & Defamation / posting attribution & PC Forensic Image & Shared family computer & CAT-DEN & Bayyinah (proof sufficiency) & 0 — Rejection & Yes & Conviction (strong corroboration) \\
C & 1432 & 2011--2012 & Privacy / reputational posting & Mobile Device Log & Communal mobile / device access & CAT-DEN & Istishab (presumption) & 0 — Rejection & Yes & Conviction (exclusive access; inconclusive denial) \\
\rowcolor{navylt}
D & 1433 & 2012--2013 & Online threat / abusive message & Account Metadata & Shared account / device access & CAT-DEN & Niyya (intent) & 0 — Rejection & Yes & Conviction (behavioural / corroborated) \\
E & 1433 & 2012--2013 & Defamation / posting attribution & IP Log (Open Wi-Fi) & Open or weakly secured Wi-Fi & SAB-TRANS & Shubha (legal doubt) & 2 — Mitigation / doubt & No & Mitigation (unsecured network) \\
\rowcolor{navylt}
F & 1434 & 2013 & Blackmail / coercive communication & Social Media Log & Shared account / device access & SAB-TRANS & Ta'zir discretion & 2 — Mitigation / doubt & No & Mitigation (ta'zir discretion) \\
G & 1434 & 2013 & Unauthorised access allegation & Router Admin Log & Household router / network & SAB-TRANS & Shubha (legal doubt) & 3 — Non-conviction & No & Non-conviction (multiple users; doubt-based) \\
\rowcolor{navylt}
H & 1434 & 2013 & Technical-report attribution dispute & Forensic Report & IP log without user-level corroboration & EVID-DISP & Yaqin (certainty) & 3 — Non-conviction & No & Non-conviction (trace gap; insufficient detail) \\
I & 1435 & 2014 & Account-use attribution dispute & Login History & Device logs without individual identification & EVID-DISP & Shubha (legal doubt) & 2 — Mitigation / doubt & No & Mitigation / non-conviction (granularity) \\
\rowcolor{navylt}
J & 1435 & 2014 & Blackmail / coercive communication & Device Ownership & Shared account / device access & MOR-REL & Tazkiyah (credibility) & 0 — Rejection & Yes & Conviction (physical evidence; ambiguity resolved) \\
K & 1435 & 2014 & Privacy / reputational posting & Account Logs & Communal mobile / device access & MOR-REL & Niyya (intent) & 2 — Mitigation / doubt & No & Mitigation (character and intent) \\
\rowcolor{navylt}
L & 1435 & 2014 & Hacking / framing scenario & Remote Access Log & Unknown third-party network exploitation & ALT-CAUS & Shubha (legal doubt) & 3 — Non-conviction & No & Non-conviction (alternative causation not excluded) \\
\end{tabular}\end{center}
\tnote{Every case involves shared digital infrastructure (\texttt{shared\_access}=Yes). ``Uptake'' gives the level and its label; ``Conv.'' is whether a conviction was recorded.}
\end{landscape}
\section{Multidimensional Discourse Integration Model — worked application to Case~G}
The MDIM links five analytic layers from technical input to judicial output. Case~G (unauthorised-access allegation; router admin log; household network) illustrates the full mapping.
\vspace{4pt}
\begin{center}\small\begin{tabular}{@{}L{4.3cm}L{11.2cm}@{}}
\hrow \hc{MDIM layer} & \hc{Case~G coding / analytic output} \\
\textbf{Technical input} & IP address identifies the subscriber's DSL line at a specific timestamp. \\
\rowcolor{navylt}
\textbf{Forensic-pragmatic layer} & Defendant uses SAB-TRANS: ``the modem is in the majlis (guest room); many visitors and my brothers have the password.'' \\
\textbf{Evidentiary evaluation} & Court notes that the forensic report confirms the source but cannot confirm the MAC address of the specific handset used. \\
\rowcolor{navylt}
\textbf{Islamic reasoning layer} & Judge invokes al-hudud tudra'u bi-l-shubuhat (punishments are warded off by doubts), applied here to ta'zir. \\
\textbf{Output} & Non-conviction for the specific charge: the perpetrator cannot be individualised among a group of potential users. \\
\end{tabular}\end{center}
\section{Inter-coder reliability — segment-level codings (n = 50)}
Coder~A is the adjudicated coding. Disagreements ($\times$) occur at S06, S31, and S42; all other segments agree.
\vspace{4pt}
\begin{center}\small\begin{tabular}{@{}C{0.8cm}C{0.7cm}L{2.0cm}L{2.0cm}C{0.7cm}!{\color{rulegray}\vrule}C{0.8cm}C{0.7cm}L{2.0cm}L{2.0cm}C{0.7cm}@{}}
\hrow \hc{Seg} & \hc{Case} & \hc{Coder A} & \hc{Coder B} & \hc{Agr.} & \hc{Seg} & \hc{Case} & \hc{Coder A} & \hc{Coder B} & \hc{Agr.} \\
S01 & A & CAT-DEN & CAT-DEN & \checkmark & S26 & F & SAB-TRANS & SAB-TRANS & \checkmark \\
\rowcolor{navylt}
S02 & A & CAT-DEN & CAT-DEN & \checkmark & S27 & G & SAB-TRANS & SAB-TRANS & \checkmark \\
S03 & A & CAT-DEN & CAT-DEN & \checkmark & S28 & G & SAB-TRANS & SAB-TRANS & \checkmark \\
\rowcolor{navylt}
S04 & A & CAT-DEN & CAT-DEN & \checkmark & S29 & G & EVID-DISP & EVID-DISP & \checkmark \\
S05 & B & CAT-DEN & CAT-DEN & \checkmark & S30 & G & EVID-DISP & EVID-DISP & \checkmark \\
\rowcolor{navylt}
S06 & B & CAT-DEN & SAB-TRANS & $\times$ & S31 & H & EVID-DISP & CAT-DEN & $\times$ \\
S07 & B & CAT-DEN & CAT-DEN & \checkmark & S32 & H & EVID-DISP & EVID-DISP & \checkmark \\
\rowcolor{navylt}
S08 & B & CAT-DEN & CAT-DEN & \checkmark & S33 & H & EVID-DISP & EVID-DISP & \checkmark \\
S09 & B & CAT-DEN & CAT-DEN & \checkmark & S34 & H & EVID-DISP & EVID-DISP & \checkmark \\
\rowcolor{navylt}
S10 & C & CAT-DEN & CAT-DEN & \checkmark & S35 & I & EVID-DISP & EVID-DISP & \checkmark \\
S11 & C & CAT-DEN & CAT-DEN & \checkmark & S36 & I & EVID-DISP & EVID-DISP & \checkmark \\
\rowcolor{navylt}
S12 & C & CAT-DEN & CAT-DEN & \checkmark & S37 & I & EVID-DISP & EVID-DISP & \checkmark \\
S13 & C & CAT-DEN & CAT-DEN & \checkmark & S38 & I & MOR-REL & MOR-REL & \checkmark \\
\rowcolor{navylt}
S14 & D & SAB-TRANS & SAB-TRANS & \checkmark & S39 & J & MOR-REL & MOR-REL & \checkmark \\
S15 & D & SAB-TRANS & SAB-TRANS & \checkmark & S40 & J & MOR-REL & MOR-REL & \checkmark \\
\rowcolor{navylt}
S16 & D & SAB-TRANS & SAB-TRANS & \checkmark & S41 & J & MOR-REL & MOR-REL & \checkmark \\
S17 & D & SAB-TRANS & SAB-TRANS & \checkmark & S42 & J & MOR-REL & SAB-TRANS & $\times$ \\
\rowcolor{navylt}
S18 & E & SAB-TRANS & SAB-TRANS & \checkmark & S43 & K & MOR-REL & MOR-REL & \checkmark \\
S19 & E & SAB-TRANS & SAB-TRANS & \checkmark & S44 & K & MOR-REL & MOR-REL & \checkmark \\
\rowcolor{navylt}
S20 & E & SAB-TRANS & SAB-TRANS & \checkmark & S45 & K & MOR-REL & MOR-REL & \checkmark \\
S21 & E & SAB-TRANS & SAB-TRANS & \checkmark & S46 & K & ALT-CAUS & ALT-CAUS & \checkmark \\
\rowcolor{navylt}
S22 & E & SAB-TRANS & SAB-TRANS & \checkmark & S47 & L & ALT-CAUS & ALT-CAUS & \checkmark \\
S23 & F & SAB-TRANS & SAB-TRANS & \checkmark & S48 & L & ALT-CAUS & ALT-CAUS & \checkmark \\
\rowcolor{navylt}
S24 & F & SAB-TRANS & SAB-TRANS & \checkmark & S49 & L & ALT-CAUS & ALT-CAUS & \checkmark \\
S25 & F & SAB-TRANS & SAB-TRANS & \checkmark & S50 & L & ALT-CAUS & ALT-CAUS & \checkmark \\
\end{tabular}\end{center}
\section{Cohen's $\kappa$ — inter-coder reliability computation}
\vspace{2pt}
\subsection*{Confusion matrix (rows = Coder A, columns = Coder B)}
\begin{center}\small\begin{tabular}{@{}lccccc|c@{}}
\hrow \hc{A \textbackslash\ B} & \hc{CAT-DEN} & \hc{SAB-TRANS} & \hc{EVID-DISP} & \hc{MOR-REL} & \hc{ALT-CAUS} & \hc{Row} \\
\textbf{CAT-DEN} & 12 & 1 & 0 & 0 & 0 & 13 \\
\rowcolor{navylt}
\textbf{SAB-TRANS} & 0 & 15 & 0 & 0 & 0 & 15 \\
\textbf{EVID-DISP} & 1 & 0 & 8 & 0 & 0 & 9 \\
\rowcolor{navylt}
\textbf{MOR-REL} & 0 & 1 & 0 & 7 & 0 & 8 \\
\textbf{ALT-CAUS} & 0 & 0 & 0 & 0 & 5 & 5 \\
\midrule
\textbf{Col} & 13 & 17 & 8 & 7 & 5 & 50 \\
\end{tabular}\end{center}
\subsection*{Statistics}
\begin{center}\small\begin{tabular}{@{}L{6.2cm}L{6.0cm}@{}}
\hrow \hc{Quantity} & \hc{Value} \\
Observed agreement ($p_o$) & 0.9400 \\
\rowcolor{navylt}
Expected agreement ($p_e$) & 0.2308 \\
Cohen's $\kappa$ & \textbf{0.9220} \\
\rowcolor{navylt}
Agreements / total & 47 / 50 \\
Disagreements & 3 (S06, S31, S42) \\
\rowcolor{navylt}
Interpretation & Almost perfect (Landis \& Koch 1977) \\
\end{tabular}\end{center}
\tnote{$\kappa = (p_o - p_e)/(1 - p_e)$. Marginals: Coder~A 13, 15, 9, 8, 5; Coder~B 13, 17, 8, 7, 5 across CAT-DEN, SAB-TRANS, EVID-DISP, MOR-REL, ALT-CAUS.}
\section{Descriptive distributions (N = 12)}
\subsection*{Primary defence strategy}
\begin{center}\small\begin{tabular}{@{}L{8.6cm}C{3.0cm}C{2.4cm}@{}}
\hrow \hc{Primary defence strategy} & \hc{Frequency} & \hc{Share} \\
CAT-DEN (Categorical denial) & 3 & 25.0\% \\
\rowcolor{navylt}
SAB-TRANS (Shared-access blame transfer) & 4 & 33.3\% \\
EVID-DISP (Evidential displacement) & 2 & 16.7\% \\
\rowcolor{navylt}
MOR-REL (Moral-religious credibility work) & 2 & 16.7\% \\
ALT-CAUS (Alternative-causation narrative) & 1 & 8.3\% \\
\midrule \textbf{Total} & \textbf{12} & \textbf{100.0\%} \\
\end{tabular}\end{center}
\subsection*{Judicial uptake and convictions}
\begin{center}\small\begin{tabular}{@{}L{8.6cm}C{3.0cm}C{2.4cm}@{}}
\hrow \hc{Judicial uptake level} & \hc{Frequency} & \hc{Share} \\
0 — Rejection & 4 & 33.3\% \\
\rowcolor{navylt}
1 — Procedural recognition & 0 & 0.0\% \\
2 — Mitigation / doubt & 5 & 41.7\% \\
\rowcolor{navylt}
3 — Non-conviction & 3 & 25.0\% \\
\midrule \textbf{Convictions recorded} & \textbf{4} & \textbf{33.3\%} \\
\end{tabular}\end{center}
\subsection*{Primary $\times$ secondary strategy co-occurrence}
\begin{center}\small\begin{tabular}{@{}lccccc@{}}
\hrow \hc{Primary \textbackslash\ Secondary} & \hc{CAT-DEN} & \hc{SAB-TRANS} & \hc{EVID-DISP} & \hc{MOR-REL} & \hc{ALT-CAUS} \\
\textbf{CAT-DEN (n=3)} & – & 1 & 0 & 1 & 1 \\
\rowcolor{navylt}
\textbf{SAB-TRANS (n=4)} & 2 & – & 2 & 1 & 0 \\
\textbf{EVID-DISP (n=2)} & 1 & 1 & – & 0 & 0 \\
\rowcolor{navylt}
\textbf{MOR-REL (n=2)} & 1 & 1 & 0 & – & 0 \\
\textbf{ALT-CAUS (n=1)} & 0 & 0 & 1 & 0 & – \\
\end{tabular}\end{center}
\tnote{Descriptive counts of secondary strategies observed alongside each primary strategy; diagonal is not applicable. Layered denial is common, so row sums may exceed the primary-strategy frequency.}
\end{document}