Overcoming
hate

whoisallah.org.uk · Common Ground · research stream K2 · 8 Nov 2026 · Pixeleyed Pictures Ltd Labels: [V] read at source this session (or title/authors confirmed via Crossref); [S] secondary or not re-read. Builds on R12 "Evidence for cohesion design" (contact theory, Mo Salah, More in Common): this file adds the practices, the plain-English explanations and one action each.

---

1. Meet people (direct contact)

In plain English. Prejudice shrinks when people actually meet, especially when they meet as equals, work on something together, and the meeting has the blessing of people they respect (a head teacher, an imam, a vicar, a rabbi). It works less by teaching facts than by lowering nerves and building fellow-feeling: once you have shared a meal with someone, "them" becomes a name.

Evidence.

Pettigrew & Tropp 2006, "A meta-analytic test of intergroup contact theory", *Journal of Personality and Social Psychology* 90(5): 515 studies, about 250,000 people in 38 nations; contact reliably goes with lower prejudice, and works beyond race and ethnicity. https://doi.org/10.1037/0022-3514.90.5.751 [V title; figures V via R12]

Pettigrew & Tropp 2008, "How does intergroup contact reduce prejudice? Meta-analytic tests of three mediators", *EJSP* 38: reducing anxiety and building empathy matter more than gaining knowledge. https://doi.org/10.1002/ejsp.504 [V]

Lemmer & Wagner 2015, *EJSP* 45: meta-analysis of real-world (not lab) contact interventions; both direct and indirect contact reduce ethnic prejudice, and effects persist. https://doi.org/10.1002/ejsp.2079 [V title; finding S]

Caveat: Paluck, Green & Green, "The contact hypothesis re-evaluated", *Behavioural Public Policy* (2018 online; 2019 print): stronger studies show smaller effects. Measure, don't assume. https://doi.org/10.1017/bpp.2018.25 [V title]

UK schools: University of Kent evaluation of The Linking Network's Schools Linking, 2024/25 (Cameron, Kemp & Horn, 2026), 664 teacher responses from 200+ schools: teachers who thought pupils understood the lives of children from different backgrounds rose from 15.9% to 78.1%; 86.7% strongly agreed pupils were more comfortable mixing by the end of the year. Teacher-reported, not a controlled trial. https://kar.kent.ac.uk/114952/ [V, executive summary read]

One action today. Pick one date from the calendar (Mitzvah Day on Sun 22 Nov 2026, or an Open Iftar in Ramadan 2027) and say yes to it, with a friend.

---

2. Know someone who knows someone (extended, imagined and media contact)

In plain English. You don't always have to meet first. Knowing that a friend has a Muslim or Jewish friend softens attitudes ("extended contact"). Even imagining a friendly conversation beforehand helps, especially for people with little chance to meet. And a familiar, likeable person on screen who visibly lives their faith works like a neighbour you see every week.

Evidence.

Wright, Aron, McLaughlin-Volpe & Ropp 1997, "The extended contact effect", *JPSP* 73(1). https://doi.org/10.1037/0022-3514.73.1.73 [V title]

Crisp & Turner 2009, "Can imagined interactions produce positive perceptions?", *American Psychologist* 64(4). https://doi.org/10.1037/a0014718 [V title]

Schiappa, Gregg & Hewes 2005, "The Parasocial Contact Hypothesis", *Communication Monographs* 72(1). https://doi.org/10.1080/0363775052000342544 [V title; corrects the DOI pattern used elsewhere]

Banas, Bessarabova & Massey 2020, "Meta-Analysis on Mediated Contact and Prejudice", *Human Communication Research* 46. https://doi.org/10.1093/hcr/hqaa004 [V title]

The Mo Salah effect: Alrababa'h, Marble, Mousa & Siegel 2021, "Can Exposure to Celebrities Reduce Prejudice? The Effect of Mohamed Salah on Islamophobic Behaviors and Attitudes", *APSR* 115(4): after Salah joined Liverpool, Merseyside hate crimes fell about 16% against a synthetic comparison, and Liverpool fans' anti-Muslim tweets roughly halved. https://doi.org/10.1017/S0003055421000423 [V title; figures V via R12]

One action today. Share one real story of a friendship across faiths (yours, or one from the site's "Neighbours" films) with someone who has never had that friendship.

---

3. Step in safely (bystander intervention: the 5 Ds)

In plain English. Most people want to help when they see someone harassed but freeze, or assume someone else will act. Having a simple plan in your head makes it far more likely you'll do something, and none of the options requires confronting anyone. The 5 Ds: Distract (interrupt: ask the time, drop something, start a chat with the person targeted); Delegate (get help from a driver, shop staff, security, or call 999 if there is danger); Document (if someone is already helping and it's safe, record, then give the footage to the victim; never post it without their permission); Delay (afterwards, check on the person: "Are you OK? That wasn't right"); Direct (speak up calmly and briefly, only if safe). Your safety and theirs come first.

Evidence and UK guidance.

Fischer et al. 2011, "The bystander-effect: A meta-analytic review on bystander intervention in dangerous and non-dangerous emergencies", *Psychological Bulletin* 137(4): the "someone else will help" effect is real but weaker in clearly dangerous situations, where people are more likely to act. https://doi.org/10.1037/a0023304 [V title; finding S]

The 5 Ds were popularised by Hollaback!, now Right To Be (righttobe.org; the site returned 403 to our scripts) [S].

UK versions: Neighbourhood Watch (Our Watch), "The 5Ds of Bystander Intervention", including "never post it online or use it without their permission" and "your safety and that of the person being harassed come first". https://www.ourwatch.org.uk/5Ds [V]

Avon and Somerset Police, "Be an upstander, not a bystander": Direct, Distract, Delegate, Defer ("take action later … or making an online crime report") and Document, and points to bystander training. https://www.avonandsomerset.police.uk/crime-prevention-advice/protect-yourself-and-others/be-an-upstander/ [V]. Conflict to note: police use "Defer" where Right To Be uses "Delay"; same idea (act afterwards).

Report afterwards: True Vision, https://www.report-it.org.uk/ (police-funded; all strands) [V, R12]; CST for antisemitism; British Muslim Trust for anti-Muslim hate (R12).

One action today. Learn the 5 Ds now (two minutes), and save report-it.org.uk in your phone.

---

4. Correct myths the way that works (the Debunking Handbook order)

In plain English. Repeating a myth to knock it down can make it stickier. The fix is to lead with the truth, mention the myth only once with a warning, explain why it's wrong (the trick it uses), and finish with the truth again: a "truth sandwich". Better still, warn people in advance about the tricks ("prebunking"), so they spot them themselves.

Evidence.

Lewandowsky, Cook, Ecker et al., *The Debunking Handbook 2020*: Fact (state the truth first, if it's clear and sticky) → Myth (warn that a myth is coming; mention it only once) → Fallacy (explain how it misleads) → Fact (repeat the truth, ideally with an alternative explanation). https://skepticalscience.com/debunking-handbook-2020-downloads-translations.html [V page loads; order S from the handbook, widely reproduced]

Lewandowsky, Ecker, Seifert, Schwarz & Cook 2012, "Misinformation and Its Correction", *Psychological Science in the Public Interest* 13(3). https://doi.org/10.1177/1529100612451018 [V title]

Prebunking: Roozenbeek, van der Linden, Goldberg, Rathje & Lewandowsky 2022, "Psychological inoculation improves resilience against misinformation on social media", *Science Advances* 8(34): short videos exposing manipulation techniques (scapegoating, false dichotomies) improved people's ability to spot them, including in a YouTube field study. https://doi.org/10.1126/sciadv.abo6254 [V title; finding S]

For the site. Every "myth" card on whoisallah should use the order: Fact → (one-line) Myth → why it's wrong → Fact. Example: *Fact:* Arabic-speaking Christians and Jews also say "Allah" for God. *Myth:* "Allah is a different god." *Why it misleads:* it treats a word as if it were a name for a separate being. *Fact:* "Allah" is the Arabic word for God, the God of Abraham. (Cross-check wording against R1 and R17.)

One action today. Next time you correct something online, start and end with the true sentence, and mention the false one only once.

---

5. Listen first (deep canvassing and perspective-taking)

In plain English. Arguments rarely change minds; conversations sometimes do. In "deep canvassing", a volunteer listens without judging, asks the other person about a time they needed compassion, and shares a story of their own. Ten-minute conversations like this have produced lasting change. Similarly, briefly stepping into another person's shoes (for example, through a role-play game) can reduce prejudice for weeks.

Evidence.

Broockman & Kalla 2016, "Durably reducing transphobia: A field experiment on door-to-door canvassing", *Science* 352(6282): a single 10-minute perspective-taking conversation reduced prejudice, with effects lasting at least three months. https://doi.org/10.1126/science.aad9713 [V title; finding S]

Kalla & Broockman 2020, "Reducing Exclusionary Attitudes through Interpersonal Conversation: Evidence from Three Field Experiments", *APSR* 114(2): the non-judgemental exchange of narratives was the active ingredient (towards immigrants and trans people). https://doi.org/10.1017/S0003055419000923 [V title; finding S]

Simonovits, Kézdi & Kardos, "Seeing the World Through the Other's Eye: An Online Intervention Reducing Ethnic Prejudice", *APSR* (online 2017; print 2018): an online perspective-taking game reduced anti-Roma prejudice among Hungarian young adults, lasting at least a month. https://doi.org/10.1017/S0003055417000478 [V title; finding S]

Caution: these are US and Hungarian studies on other groups. They support the *method* (story exchange, non-judgemental listening), not a promised size of effect for religious prejudice in the UK.

For the site. "Same Word nights" should end with a story exchange ("Tell us about a time a stranger was kind to you"), not a debate.

One action today. Ask one person whose view on Muslims, Jews or Christians differs from yours: "What's your experience been?" Then listen without correcting for five minutes.

---

6. What UK programmes teach

| Programme | What it teaches | Evidence it publishes | Source | |---|---|---|---| | The Linking Network (1163901) | Four questions for pupils: Who am I? Who are we? Where do we live? How do we all live together? [S, not re-read] Year-long class linking with a face-to-face day | Kent evaluation 2024/25 (above) [V]; earlier NFER/DfE evaluation of the Schools Linking Network (DFE-RR090, 2011) [S, not read] | https://thelinkingnetwork.org.uk/ | | The Faith & Belief Forum (1173369) | School linking, speakers of different beliefs, university programmes: meet the person, ask questions | 2024–25: "linking 60 classes in London and Birmingham"; resources downloaded "over 13,800 times" [V, via R18] | https://faithbeliefforum.org/ | | HOPE Unlimited Charitable Trust (1013880) | Anti-hate education and research (the charity behind HOPE not hate's education work) | Not read this session [S] | https://hopeunlimited.org.uk/ | | Tim Parry Johnathan Ball Foundation for Peace (1048990) | Non-violence for young people, built on the Warrington bombing story; support for survivors of terrorism | Not read [S] | https://timparryjohnathanballfoundation.org.uk/ | | Anne Frank Trust UK (1003279) | Challenging prejudice through Anne Frank's diary; young ambassadors | Not read this session [S] | https://annefrank.org.uk/ | | Holocaust Educational Trust (1092892) | Lessons from Auschwitz: students become ambassadors and teach others | Not read [S] | https://www.het.org.uk/lessons-from-auschwitz-programme |

Common thread: they combine contact (meet), story (hear a real person) and action (do something together), and they make the first meeting feel safe. That is the shape for whoisallah's "from the screen to the street": watch → meet → do.