Tree Person House Test

What the HTP Actually Measures (And What It Doesn't)

The House-Tree-Person test is a projective drawing exercise where someone sketches a house, a tree, and a human figure on blank paper. Karen Machover popularized it in the 1950s. The assumption behind it is that people project unconscious material into their drawings, and a trained analyst can read that material back. Most psychologists I've worked with treat it as a supplementary tool, not a standalone diagnostic. It's useful in clinical intake when you need to break the ice with a reluctant client, or when verbal self-report is compromised by resistance or developmental limitations. You need three sheets of plain white paper, size 8.5 by 11 inches, a standard pencil, and an eraser. Don't give them a pen. The instruction is simple: "Draw a house." Then "Draw a person." Then "Draw a tree." No time limit. No subject specification. Let them pick their own person, their own tree species, their own style of house. The lack of constraint is the whole point. I've seen people flip these orders themselves. Someone will draw the person first because they're anxious about being evaluated, or they'll draw the tree last because houses feel safer to them psychologically. Note the order. It matters more than most guides admit.

While they're drawing, don't talk. Don't offer praise. Don't ask leading questions like "Does that look happy?" Record anything notable but keep it minimal: how long they spent on each drawing, whether they erased heavily, if they asked what a proper drawing should look like. That last one tells you something about their relationship to authority and judgment before you even look at the finished product. After all three are complete, ask them to tell you about each one. "What's happening in this picture?" is better than "What does this mean?" Let them say what they want. People will reveal stuff in the post-drawing interview that contradicts or complicates what the symbols alone suggest. I once had a client draw a house with no door and I almost wrote that off as isolation symbolism. When I asked about it, they said the door was just on the other side and they couldn't be bothered to draw it. Fair enough. Context matters. The whole process takes roughly 15 to 30 minutes depending on the subject. Adults usually move faster. Children take longer and often need the prompt repeated once or twice because they don't always understand what's being asked.

Common scoring frameworks include Buck's Complete Projective Technique, which catalogs specific features like window size, roof detail, foundation presence, and root visibility. Hammer's expressive drawing analysis focuses more on spatial placement and pressure. There's no single authoritative system and that's one of the biggest problems with this test. You pick a framework, you commit to it, and you acknowledge that different examiners using different frameworks can reach opposite conclusions from the same drawing. Here's something beginners miss: the tree isn't just about the person's relationship with their environment. In most interpretive systems, the roots represent unconscious drives or grounding, the trunk is ego strength, the branches are outreach and aspiration, and the canopy is the conscious self. But I've seen people draw trees with enormous canopies and tiny trunks and clinicians immediately label it as grandiosity. Sometimes the person was just drawing an oak tree and oaks have big canopies. Specifying the tree type during the post-interview can save you from a wrong reading half the time. Another thing nobody emphasizes enough: paper placement and body posture during the drawing are data points. A subject who consistently draws everything in the upper left corner tends to be more introspective or nostalgic. Upper right is more future-oriented. Heavy smudging on one side often correlates with anxiety about that life domain. This isn't hard science, but it's consistent enough to be worth tracking across sessions.

Where the Tree Person House Test Falls Apart

Reliability is the core problem. Two trained clinicians looking at the same set of three drawings can produce different interpretations, sometimes wildly different ones. The test lacks standardized scoring criteria that all practitioners agree on. You'll find reliability coefficients in the literature ranging from poor to moderate depending on which feature you're measuring. Some features like figure size relative to the page have okay inter-rater agreement. Features like roof shape or fence presence are basically interpretive free-foralls. Cultural bias is another real issue. The house category assumes a residential structure that looks a certain way. People from cultures where housing norms differ, or from socioeconomic backgrounds where permanent structures are uncommon, may draw something that looks "deficient" by Western middle-class standards when it's just a culturally normal representation. I had a client draw a tent-like structure for the house and the supervisor wanted to code it as instability. The client grew up in a family that moved frequently and the tent was literally what a house meant to them. That reading would have been nonsense.

The test also doesn't account well for motor or visual-spatial disabilities. A person with dysgraphia or low vision will produce drawings that look pathological on paper but reflect physical limitation rather than psychological conflict. I learned this the hard way when I nearly flagged a drawing as severely regressive because the figures were unusually small and poorly detailed. The person had cerebral palsy affecting fine motor control. The interpretation would have been embarrassing and wrong. For adults being assessed for employment or legal purposes, the HTP has almost no forensic value. It doesn't hold up under scrutiny in those contexts. If someone needs a test that survives cross-examination, use the MMPI or the PAI instead. The HTP belongs in therapeutic settings where it's part of a broader clinical picture, not as evidence. If you want to download scoring guidelines or reference material, most of the publicly available PDFs are either out-of-copyright reproductions of Machover's original work or proprietary systems sold through psychological publishers. The free ones online tend to be simplified cheat sheets. For serious use, you'd want to get one of the published manuals directly from the source publisher rather than relying on random websites. The interpretations matter too much to trust a blog post.

The test still has a place. It's fast, cheap, and the drawing act itself can be therapeutic for clients who struggle to articulate feelings verbally. Children respond to it better than almost any other projective instrument. But treat it like a conversation starter, not a verdict. The drawings are a map, not the territory.