A separate published SD4 subscale
Read the seven-item Sadism mean with its actual wording, source and limits. No diagnosis, type or population rank.
Seven numeric ratings required; raw sum / 7. Any unscored response means no personal mean for this subscale.
A mixed seven-item content set
The SD4 sadism subscale includes excitement at a fist-fight, enjoyment of violent films and games, amusement at another person’s fall, violent sports, a belief that some people deserve suffering, mean online comments for fun and knowing how to hurt someone with words. These contents are related within the published research instrument but are not identical actions. A higher mean records stronger agreement across the exact seven statements; it does not prove enjoyment of every kind of suffering or establish any specific harmful act.
Fictional media does not establish real-world harm
Enjoying a violent fictional game is not equivalent to hurting a real person. Watching a sport also differs from initiating aggression, and knowing that words can hurt differs from having used them in a particular exchange. Because these different statements contribute to one mean, the item pattern matters. Two people with the same number may have endorsed entirely different content. The report cannot infer what an individual viewer or player would do outside the media setting.
A score is not a moral or safety verdict
The authors’ research term is not a personal identity assigned here. This subscale does not diagnose a sexual interest or disorder, identify an abuser, decide whether a relationship is safe or rank people as good and bad. No cutoff is supplied. A low mean does not certify kindness, while a high mean does not by itself establish intent, severity or responsibility for an event. Use concrete evidence and appropriate support for actual concerns rather than asking this average to settle them.
Distress, provocation and interpretation differ
One statement refers to mean social-media comments made for amusement, while another is a broad belief about deserved suffering. Remembering an instance does not tell the website what another person felt or how an interaction developed. The scale also does not establish why you selected an answer. An interpretation of an insulting exchange, humour or competition may differ across settings and languages. These limitations remain even when every question has a numeric response and the arithmetic is complete.
The fourth subscale has its own source identity
SD4 adds a sadism subscale to its other three research subscales, each with seven items. That does not make the fourth mean a supplement to an unrelated Mini score, a humour preference score or a competitor’s four-type quiz. The published 28 English items, five agreement anchors and four direct scoring keys are retained here. The optional unscored response and rule withholding a mean when any of its seven ratings is missing are website extensions, not an author-validated missing-data procedure.
Notice one interpretation without testing harm
If further reflection is useful, consider what you understood by one media, sport or online-comment statement. You do not need to recreate a situation, provoke anyone, seek proof from another person or disclose private incidents. Distinguish an imagined scene from a real action and from its consequences. No behavioural experiment is suggested. You can leave the part unscored, read any of the four public guides or finish without deciding what the research label says about you.
Source, language and the limits of the version
This page follows the 28 English statements and five agreement anchors published on coauthor Erin Buckels’ SD4 project page, crosschecked against the instrument distributed by Delroy Paulhus. Paulhus lists SD4 among measures in the public domain available without permission. The cited article is Paulhus, Buckels, Trapnell and Jones, published in 2021. Our Chinese wording is a working translation, not a validated Chinese form. Publication of the original instrument does not establish validity, norms or accuracy for this website’s users or this translation.
Seven direct ratings and an unrounded mean
Use 1 Strongly Disagree, 2 Disagree, 3 Neutral, 4 Agree and 5 Strongly Agree. Three is the numeric neutral response, not missing information. All 28 items are direct, with no reverse scoring. For each subscale, seven numeric ratings produce their sum divided by seven, displayed to two decimals from the unrounded sum. One, two, three, four, five, three and three sum to 21 and give 3.00. Any unscored response withholds that subscale’s mean; nothing is imputed. No total, dominant type, percentile or cutoff is calculated.