Scoring the Drawing Test When It's Not Going Smoothly
The Goodenough-Harris Drawing Test is a projective drawing assessment that uses a child's drawing of a person to estimate cognitive ability. It was originally developed by Ruth Goodenough in 1926 and later revised to include Harris's contributions in 1963. The test presents three drawing instructions to the child — draw a picture of yourself, draw a picture of a boy, and draw a picture of a girl. A trained examiner scores the drawings based on 25 specific items across the three figures, producing a Mental Age score and a corresponding IQ score. Most psychologists and school counselors who encounter it do so through the manual, which lays out the administration and scoring procedures. The manual covers what most people need: the standard scoring criteria for each of the 25 items, the norms by age and gender, and the administration guidelines. What the manual doesn't always spell out clearly is how to handle situations where the child draws something unexpected or partially completes a figure. That's where practice matters more than reading instructions carefully.
How to Access the Goodenough Harris Drawing Test Manual
The manual is published by Pearson as part of the Harvey Screened Individual Mental Abilities Scale materials. It's available through academic publishers and most testing supply companies. If you're a student or researcher on a budget, check your university library's test collection first. Many campuses hold physical copies of the manual that you can consult on-site, which is often the fastest route compared to waiting for interlibrary loan. Digital copies sometimes circulate through institutional repositories, but availability depends heavily on your affiliation and region. The core content of the manual includes: the 25 scoring items broken down by figure (self, boy, girl), the scoring rubric for each item, the norm tables for ages 6 through 14, instructions for presentation and probing, and notes on interpretation and limitations. You'll also find the standardization sample details, which gives you some idea of what the reliability actually looks like across different populations.
What Actually Happens When You Administer It
You hand the child a pencil and blank white paper. You give them the three instructions one at a time. You watch them draw. You score according to the manual. That's the surface-level version. In practice, the child might refuse to draw a particular figure, draw something very crude, add excessive detail to one area and ignore another, or produce drawings that are too small or placed oddly on the page. Scoring is mostly binary — the feature is present or it isn't. You award one point for each correctly rendered item and zero for incorrect or missing features. The 25 items cover head shape, eyes, eyebrows, mouth, ears, nose, neck, torso, arms, hands, legs, feet, clothing, fingers, toes, and body proportions. Each of the three figures contributes to the total, and the total raw score converts to a Mental Age using the norm tables. From there you can derive an IQ score if you have the child's chronological age. One thing the manual treats lightly is the scoring ambiguity around proportionality. Does a figure where the head is enormous relative to the body get a zero on the body proportion item, or does it depend on whether the body is even sketched in? I've seen scorers disagree on this. The manual says proportionate body parts earn the point, but it doesn't define proportionate with enough precision for everyone. During my calibration sessions with other examiners, I learned to apply a consistent rule: if the torso is clearly visible and discernibly smaller than the head by more than a factor of two, the proportionality point is missed regardless of whether it's intentionally drawn that way. Consistency across scorers matters more than chasing the letter of the manual.
Get the Full Details
Common Pitfalls That Beginners Miss
The first issue most people run into is timing. Children take wildly different amounts of time to complete each figure. Some finish all three in under five minutes. Others need fifteen or twenty. The manual doesn't set a strict time limit per figure, which sounds generous but becomes problematic when you're working with a group of examiners who need to produce scores quickly. I once had a junior examiner stop a child mid-drawing after exactly ten minutes, claiming the manual said ten minutes. It didn't. That child's third figure was incomplete, and the resulting score was artificially low. Let the child work at their own pace. The instructions are clear about this, but it's easy to lose sight of when you're trying to stay on schedule. Another issue is probing. The manual allows you to ask the child to explain parts of the drawing if they're ambiguous. But the wording of your probe matters. Asking "What is that?" about a blank space near the figure can lead the child to invent features that weren't actually drawn. I prefer neutral probes like "Tell me about this part of your drawing." It reduces the chance of the child conforming to what they think you want to hear. This is a small difference but it affects scoring accuracy, especially on borderline cases. A more subtle problem involves scoring the clothing item. The manual awards a point for any recognizable clothing detail — buttons, shoes, sleeves, collars. But some children draw clothing lines that are purely decorative and don't correspond to actual garments. A child who adds stripes to the torso without any indication of a shirt or dress shouldn't automatically get the point. I learned this through trial and error when reviewing my own scores against another examiner's. We scored the same set of drawings differently on clothing in about 30% of cases. After comparing notes, I tightened my criteria to require a discernible garment outline before awarding the clothing point.
When the Test Doesn't Work
The Goodenough-Harris Drawing Test has real limitations that the manual mentions but doesn't emphasize enough for anyone making high-stakes decisions. The test was standardized on a largely white, middle-class sample from the 1960s. Its norm tables may not apply fairly to children from different cultural backgrounds, children with limited exposure to drawing activities, or children who have motor coordination difficulties. A child with dyspraxia or cerebral palsy will likely score lower on proportionality and detail items not because of cognitive ability but because of motor execution. The manual acknowledges this but the scoring system itself doesn't adjust for it. The test also has modest reliability. Test-retest reliability coefficients typically fall in the .70 to .80 range, which is acceptable for screening but insufficient for individual diagnostic decision-making. A child who scores an IQ of 95 on one administration might score 80 or 110 on a retest two weeks later simply due to test-retest variability. I've seen counselors treat a single Drawing Test score as definitive when it shouldn't be treated as anything more than a rough estimate. If you need a more reliable measure of cognitive ability for individual children, the Woodcock-Johnson Tests of Cognitive Abilities or the WISC-V are better choices. They take longer to administer but produce scores with substantially better psychometric properties. The Drawing Test still has its place — it's quick, inexpensive, and can be useful as a screening tool or when you need a nonverbal estimate of cognitive functioning in a context where verbal testing isn't feasible. But it shouldn't be the only tool in the room.
Practical Scoring Workflow
Here's how I approach scoring when I have a stack of drawings to evaluate. First, I review the manual's scoring sheet and have it open on my desk. I score each figure independently before moving to the next. I note any ambiguities directly on the scoring sheet rather than trying to hold them in my head. For ambiguous items, I apply the same rule consistently across all drawings — the proportional rule I mentioned earlier, the clothing rule, the hand detail rule. After scoring all three figures, I sum the points and look up the Mental Age in the norm tables. I double-check the conversion, especially near boundary ages where a single point can shift the result by several months. I also keep a record of any unusual features or behaviors during administration. The manual asks for clinical observations, and I treat that seriously. A child who drew extremely small figures, covered the page with shading, or refused to complete a figure provides information that the raw score alone doesn't capture. These observations go into the report even if they don't change the numerical score. The manual remains a practical resource for professionals who need a brief, nonverbal cognitive screening tool. It's not perfect, it hasn't aged well in every demographic dimension, and it's been largely superseded by more rigorous instruments for individual assessment. But when used appropriately — as a screen, not a diagnosis — and scored with attention to the ambiguities that only practice reveals, it still provides useful information. The key is knowing what it can and cannot tell you before you hand the pencil to the child.
