The Stanford and Harvard Hypnotizability Scales: What They Measure

If hypnosis is going to be studied scientifically, there needs to be a way to measure who responds to it and who does not. The Stanford and Harvard scales have served this purpose for over sixty years. Understanding what they measure, and what they do not, is essential for interpreting the hypnosis research literature.

The Stanford Scales

The original Stanford Hypnotic Susceptibility Scales (SHSS) were developed by Weitzenhoffer and Hilgard in 1959 [1]. They exist in several forms:

Forms A and B are parallel versions containing 12 items each, arranged in increasing difficulty. A typical session takes about an hour and involves a hypnotic induction followed by standardized suggestions. The person’s score is the number of items they pass.

Form C is the most commonly used research version. Its items include:

• Easy items: Hand lowering (arm becomes heavy), moving hands apart, mosquito hallucination
• Moderate items: Taste hallucination, arm rigidity, dream suggestion
• Difficult items: Age regression, arm immobilization, anosmia (inability to smell), negative visual hallucination, post-hypnotic amnesia

Each item has specific behavioral criteria for passing. For example, arm lowering is passed if the outstretched arm drops at least six inches within ten seconds of the suggestion.

The Harvard Group Scale

The Harvard Group Scale of Hypnotic Susceptibility (HGSHS:A) was developed by Shor and Orne in 1962 [2]. It adapts the Stanford format for group administration:

• A standardized audio or live induction is given to a group
• Suggestions are delivered to the whole group simultaneously
• Participants score their own responses afterward

The Harvard scale is widely used for screening in research because it is efficient. However, self-scoring introduces a reliability concern: people may not accurately judge their own responses. The correlation between Harvard and Stanford scores is moderate (typically r = 0.60–0.70), suggesting they measure the same construct imperfectly.

What the scores mean

Score range (out of 12)	Classification	Approximate prevalence
0–3	Low hypnotizability	10–15%
4–8	Medium hypnotizability	60–75%
9–12	High hypnotizability	10–15%

These scores are remarkably stable over time. A 25-year retest study found a correlation of 0.71, meaning that someone who scores high in their twenties will likely score high in their fifties [3].

Strengths of these scales

• Standardization. The procedures are scripted, allowing comparisons across studies.
• Reliability. Both internal consistency and test-retest reliability are good.
• Validity. Scores predict laboratory hypnotic responding reliably.
• Historical continuity. Sixty years of data using these scales allows comparison across decades.

Limitations

Laboratory-clinical gap. The scales were developed for laboratory research and contain items, like hallucinations and amnesia, that are not typical of clinical hypnosis. Whether a high score predicts better clinical outcomes depends on the application and is not consistently established.

Content. Some items reflect mid-twentieth-century norms. The “mosquito hallucination” item, for example, has been criticized as culturally specific and potentially distressing.

Administration time. The Stanford scales require about an hour with a trained administrator, which limits their use in clinical practice.

Behavioral versus subjective. The scales primarily measure behavioral compliance. A person might experience vivid internal changes without producing the specific behavioral response the scale requires, or produce the behavior through compliance without the corresponding subjective experience.

Modern revisions have attempted to address some of these issues [4] [5].

Clinical use

Most clinical hypnosis practitioners do not routinely measure hypnotizability before treatment. Arguments for and against:

For measurement:

• Sets realistic expectations
• Helps tailor the intervention
• Identifies people unlikely to benefit from suggestion-based approaches

Against measurement:

• Adds time and cost
• May discourage people who score low
• Clinical hypnosis packages include non-suggestion elements that may help regardless of score
• The lab-clinical gap means scores may not predict clinical outcomes

This is an unresolved debate. For more on the clinical relevance of hypnotizability, see What Is Hypnotizability?.

What this means in practice

The Stanford and Harvard scales are the foundation of hypnosis measurement. They tell us that hypnotizability is real, measurable and stable. They do not tell us, with certainty, who will benefit from clinical hypnosis for any given condition. A person curious about their hypnotizability can be tested, but should know that the score is most informative for laboratory-type responding and may not predict their experience in a clinical session.

For how these individual differences affect research outcomes, see Why Comparison Groups Matter.

When to see a clinician first. Hypnotizability testing is a research and clinical assessment tool, not a treatment. If you are interested in hypnosis for a specific condition, discuss it with a qualified health professional who can advise on whether the evidence supports it for your situation.