What 938 people quitting porn said caused their relapse, and what they said helped

The survey

In 2015 the NoFap forum on Reddit let a researcher put a survey in front of members at the moment they reported a relapse¹. In three months it drew 1,182 responses, and 938 people wrote in their own words what had caused it¹. Rackham, at Brigham Young University, coded every answer one by one for a 2020 doctoral dissertation¹.

Rackham writes that very few answers gave a single cause, and that feelings often came two or three at a time¹. The dissertation's own count sits oddly with that, since the 938 answers were coded into only 999 reasons¹. Nine in ten respondents said they were male, the rest did not say, ages ran from 12 to 48 with an average of 22, and most were not religious¹. That is a narrow group. It is also a group that answered as soon as they reported the relapse, which almost no other study of porn use can say.

Feelings came first

27% of the reasons named a feeling, more than any other kind¹. Stress led, in 78 of the 267¹. Work and school were the cause of stress named most, often together with putting work off. Being alone came next with 55, then boredom with 52 and depression with 45¹.

Almost nobody named one feeling and stopped there. Only 13 reasons gave stress on its own, and only 4 gave depression on its own¹. Rackham's takeaway is that people often named two or three feelings together rather than one¹. Stressed and bored. Lonely and low. A plan written for "when I am stressed" can miss the evening when stress and an empty room turn up together.

Body states made up another 16% of the reasons¹. Nearly half of those were a physical urge, and 46 were tiredness or short sleep¹. 20 named a lack of exercise, and 18 named alcohol or drugs, most of them alcohol¹.

Alone in the room, or lonely

The word "alone" meant two different things. In 22 reasons it meant being alone in the room or the house, often in front of a computer¹. In 27 it meant feeling lonely¹. The first is about where you are, and the second about who you have. Rackham notes that the two would need different responses¹.

Reasons about other people made up 11% of the total, 115 of the 999¹. 70% of them also named a feeling, most often loneliness¹. The pairing ran the other way far less often. Reasons that blamed the mind named a feeling 11% of the time¹.

Relationships of every kind came up. A partner was named 23 times, wanting more sex 11 times and someone they were interested in 9 times, along with family, friends, classmates and coworkers¹. Rackham argues that relapse plans for porn tend to look mostly at romantic and sexual ties and may miss the rest¹. The advice to counsellors is to ask about relationships when someone names a feeling, and about feelings when someone names a relationship¹.

Looking, or stumbling on it

155 reasons, 16%, named sexual content the person had seen just before¹. 60 said they went looking for it¹. 38 came across it without looking, on Instagram, in a video game, in a music video on TV¹. 57 did not say which¹.

Some who went looking described a slow climb. A search for one revealing photo led to more, and then to porn. The forum calls this peeking, and the working notes count 14 answers about it and 19 more about edging¹. This is the most useful split in the study. Stumbling is a problem of what reaches you. Looking is a problem of what you do once the thought arrives. Rackham suggests asking which one it was before choosing a response¹.

Blaming the mind

230 reasons, 23%, blamed the mind¹. Lack of willpower and sexual thoughts or memories each made up 30% of them¹. Losing the reason to keep going made up 20%, and calling oneself weak another 20%¹.

These answers stood apart from the rest. Only 15% of them also named a feeling or a relationship¹. 20 answers gave willpower as the only reason, which Rackham notes stood out where very few answers named a single cause¹. Rackham reads these answers as harsh on the self. Many wrote of willpower as something they had lost and did not know how to get back, though some used the word to hold themselves to account for their choices¹.

Rackham sees memories as a problem specific to porn. A few people described images seen before that came back on their own, with no new search needed¹. And a short answer may leave out causes the person did not write down, a point Rackham raises about one-word answers like boredom¹. "I am weak" takes less typing than "I was tired, alone and putting off an essay".

What people said helped

The survey also asked what had been the most helpful strategy for staying off porn. Rackham coded the answers into 64 codes, then set the question aside for a later study, so the dissertation gives counts of mentions and no analysis¹. Keeping busy came up 189 times, keeping the reasons for quitting in mind 162, exercise 88 and being with people 81¹. Shutting the urge down by force came up 73 times¹.

Strategies that change the day or the room ranked high. Going outside came up 49 times, being productive 45, a set schedule 29 and getting away from the situation in the moment 26¹. Prayer came up 44 times, as often as meditation, in a group that was mostly not religious¹. 16 said nothing had helped¹.

Read the two halves together and they match. People relapsed on stress, empty time and being alone. What they said helped was a full day, other people and a reason to keep going. Nobody checked whether those who named a strategy relapsed less, so the second half is what people believed worked¹.

How they found it

• Two surveys ran inside the NoFap subreddit in 2015, with the moderators' agreement. The second opened on screen as soon as a member reported a relapse, and drew 1,182 responses in 3 months.
• 938 people answered the open question about what caused the relapse. An answer that gave two reasons counted twice, which turned 938 answers into 999 reasons. Two other researchers checked the codes along the way.
• After sorting the reasons into six kinds, later rounds of coding looked at which kinds turned up in the same answer. Feelings and people did, at least three times as often as any other pair, by the dissertation's account.
• The codebook of relapse causes in appendix D is labelled round 3, and some of its counts differ from the text's, such as 227 feeling reasons against 267 and 90 about people against 115. The relapse charts use the text's counts, and that codebook only for codes the text does not count.
• The same survey asked what had been the most helpful strategy. Those answers were coded into 64 codes under six headings and counted, but not analysed.
• Relapse meant porn, masturbation or orgasm after at least one clean day, by the forum's own rule. Only 5 of the 938 relapses did not include porn.

What it cannot tell you

• The sample was the forum's members, not a random one. Most were not in treatment, and nobody checked whether they met any diagnosis.
• The question asked what caused their "failure", the forum's own word. That may have drawn out bad feelings and left out good ones, like excitement, that can also come before a relapse.
• Each answer was a short written reply, with no follow-up questions. A one-word answer like "boredom" may hide causes the person did not write down.
• One researcher did most of the coding. A different coder could have drawn the lines between kinds differently.
• The strategy counts are mentions, not people, and the dissertation does not say how many answered. That question came right after a tick list offering exercise, calling someone, journaling, walking, meditating and praying.
• This shows what people said came before a relapse and what they believed helped. It cannot show that either one caused anything.

In P-Minus

The P-Minus urge log asks the same questions in the moment, as taps. What came first, where you were, whether anyone was with you and how you were, with stressed, bored and lonely among the choices, so being alone and feeling lonely are logged apart. "I saw something" is kept apart from "I was browsing near it". Once seven urges are logged, it groups them so you can see which ones keep coming back.

Sources

1. Rackham EL. Understanding relapse in self-perceived problematic pornography users [doctoral dissertation]. Provo (UT): Brigham Young University; 2020.

Sources checked October 2026.