You are viewing a free preview of this lesson.
Subscribe to unlock all 10 lessons in this course and every other course on LearningBro.
If attachment is an innate, evolved system, as Bowlby claimed, then its basic form should look similar wherever human infants are raised. If, instead, attachment is largely a product of local child-rearing customs, its patterns should differ markedly from one culture to another. Testing this is not a side issue — it goes to the heart of whether attachment theory describes something universal about human development or merely a Western way of raising children. The single most important piece of evidence on this question, and the prescribed classic study for the Edexcel Child Psychology topic, is van IJzendoorn and Kroonenberg's (1988) meta-analysis of Strange Situation studies across the world. This lesson works through that study in depth, distinguishes cross-cultural from intra-cultural (within-culture) variation, and uses the findings to weigh cultural relativism against the universality of attachment.
Key term: cultural relativism is the position that behaviour can only be properly understood, and judged as normal or abnormal, within the cultural context in which it occurs — implying that a measure developed in one culture may not be valid in another. Its opposite in this debate is universality: the claim that some psychological features are common to all humans regardless of culture.
This lesson addresses cross-cultural variations in attachment required by Edexcel 9PS0 — Paper 2, Topic 7: Child Psychology, centred on the classic study, van IJzendoorn and Kroonenberg (1988). As a classic study, you are expected to know its aim, procedure, findings and conclusions in detail and to be able to evaluate it. The lesson builds directly on the previous one: the study pools Strange Situation classifications, so its validity depends on the validity of that procedure, and it is the empirical test of the universality claim in Bowlby's evolutionary theory. In assessment-objective terms you should be able to describe the meta-analysis and its findings (AO1), apply the concepts of cross-cultural and intra-cultural variation and the imposed etic to explanations of cultural differences (AO2), and evaluate the study's methodology and its bearing on the relativism-versus-universality debate (AO3).
Connects to…
Before the study, two predictions were on the table. Bowlby's evolutionary account predicts universality: because the attachment system is inherited and adaptive, secure attachment — an infant confidently using a responsive caregiver as a secure base — should be the most common pattern everywhere, and the distribution of types should look broadly similar across cultures. The rival, culturally relativist prediction is that attachment patterns are learned social conventions that reflect local values about independence, closeness and the handling of strangers, and so should differ substantially between cultures. Isolated single-culture studies had begun to hint at differences — German samples seemed to show more avoidance, Japanese samples more resistance — but no single study could settle the matter, because each used a modest sample from one place and might simply be unrepresentative. What was needed was a way to pool many studies to see the pattern across cultures as a whole. That is precisely what a meta-analysis provides.
Key term: a meta-analysis is a research method that statistically combines the results of many separate studies on the same question, producing a single, larger and more reliable overall estimate — and allowing comparisons (here, between cultures) that no individual study could support.
Aim. To investigate whether attachment type, as measured by the Strange Situation, is universal across cultures or varies between them; and, crucially, to compare how much attachment classifications vary between countries with how much they vary within a single country.
Procedure.
Findings.
Conclusions. The dominance of secure attachment in every culture is consistent with Bowlby's claim that it is an innate, adaptive norm — the universal pattern his theory predicts. At the same time, the cross-cultural differences in the insecure types show that culture does shape how attachment is expressed. The large within-culture variation, however, is the study's deepest lesson: "culture" is not a uniform block, and differences in the immediate caregiving environment — sub-culture, social class, individual parenting — matter more than a country's overall label. It is therefore misleading to speak of a single national "attachment style".
The table below summarises the directions of difference reported (exact percentages varied across the pooled studies; the pattern is what to learn).
| Country / sample | Secure (B) | Insecure-avoidant (A) | Insecure-resistant (C) | Interpretation |
|---|---|---|---|---|
| USA | Majority secure | Moderate | Low–moderate | Close to the overall pattern (Ainsworth's origin) |
| Great Britain | Highest secure | Moderate | Low | Close to the "norm" |
| (West) Germany | Majority secure | Notably high | Low | Value placed on early independence/self-reliance |
| Japan | Majority secure | Very low | Notably high | Infants rarely separated from mother; separation unusually distressing |
| Israel (kibbutz) | Majority secure | Low | High | Limited exposure to strangers; the stranger episodes especially stressful |
| China | Majority secure | Insecure types more evenly split | Fewer classifications; pattern less clear-cut |
graph TD
A[Total variation in<br/>attachment classifications] --> B[Between-culture variation<br/>differences across the 8 countries]
A --> C[Within-culture variation<br/>differences between studies in the SAME country]
C -->|about 1.5x larger| B
B --> D[Culture shapes the<br/>INSECURE types]
C --> E[Individual caregiving<br/>matters more than nationality]
style C fill:#2980b9,color:#fff
style E fill:#27ae60,color:#fff
Exam tip: the within-versus-between finding is the most sophisticated point in this topic. State it precisely — within-culture variation was about 1.5 times greater than between-culture variation — and draw the correct inference: it undermines crude national stereotyping and points to individual caregiving, not nationality, as the main driver of attachment type.
The differences in the insecure types are best understood by asking how the Strange Situation itself interacts with different child-rearing practices — which is where the concept of the imposed etic becomes central.
Consider the Japanese samples, illustrated by a frequently cited individual study, Takahashi (1990). Takahashi ran the Strange Situation with Japanese infants and found that many became so distressed during the "infant alone" episode that the procedure had to be stopped early. In this sample, infants were almost never separated from their mothers in everyday life; being left alone in an unfamiliar room was therefore a far more extreme and unusual event for them than the procedure assumes. As a result, many Japanese infants appeared "insecure-resistant" — intensely distressed and hard to comfort — not because their attachments were insecure, but because the situation was abnormally stressful for them. Their behaviour on reunion may have reflected the unfamiliarity of separation rather than an insecure bond.
Now consider the (West) German samples, which showed high rates of insecure-avoidant attachment. Grossmann and colleagues suggested this reflected a cultural value placed on early independence and self-reliance: German caregivers often encouraged infants not to be "clingy" and to tolerate distance. An infant who has been raised to be independent may not seek proximity or comfort on reunion — behaviour the Strange Situation scores as avoidant — even though the underlying caregiving is not rejecting in the way Ainsworth's scheme assumes. The Israeli kibbutz samples, similarly, showed elevated resistance, plausibly because infants raised in a close communal setting had limited exposure to unfamiliar adults, so the stranger episodes were especially alarming.
The common thread is the imposed etic: the Strange Situation was designed around Western, and specifically American, assumptions about what healthy attachment looks like, and its scoring treats certain culturally specific behaviours as markers of insecurity. Exported unchanged, the tool risks mislabelling culturally normal behaviour as insecure.
It is worth being clear about why the secure-majority finding is such powerful support for universality, because this is easy to state weakly. The argument is not simply that "secure was common" but that secure attachment was the single most frequent classification in every one of eight very different societies — societies that differ in family structure, in how independence is valued, in exposure to strangers, and in who does the day-to-day caring. Cultural conventions of this kind pull in different directions across these societies, yet the same modal pattern emerged everywhere. A behaviour that survives such varied cultural pressures to remain the norm is much more plausibly an evolved, species-typical adaptation than a coincidence of eight independent social conventions all happening to converge. That is the logic that makes this study, despite its measurement problems, one of the strongest empirical pillars under Bowlby's universality claim.
Equally, the study should not be read as showing that culture is irrelevant. The Swedish and Dutch samples sat close to the overall pattern, but the German, Japanese and Israeli departures are real and patterned, not random noise — they line up sensibly with known differences in child-rearing. The honest reading is that culture modulates the expression and measured distribution of attachment around a universal secure core, which is exactly the two-part conclusion developed below.
Key term: an imposed etic occurs when a researcher applies a measurement tool or concept developed in one culture to another culture, assuming it is valid there without checking whether it is meaningful in that cultural context. It is contrasted with an emic approach, which studies behaviour from within a culture using locally meaningful concepts.
The great value of van IJzendoorn and Kroonenberg's study is that it does not simply come down on one side. Instead, it lets us hold a nuanced, two-part position:
The most defensible conclusion is therefore that attachment has a universal core (the secure-base relationship, adaptively selected) whose surface expression and measurement are culturally relative. This is a genuinely evaluative position, and reaching it — rather than declaring attachment simply "universal" or simply "cultural" — is what lifts an answer into the top band.
The meta-analytic method gives the study great statistical power, a real strength. Pooling 32 studies and over 1,900 classifications produces far more reliable aggregate estimates than any single study, and — uniquely — it allows the within-versus-between-culture comparison that is the study's most important contribution. A single study in one country simply could not reveal that within-culture variation exceeds between-culture variation. The large, combined dataset is a major methodological asset that gives the headline conclusions genuine weight.
Subscribe to continue reading
Get full access to this lesson and all 10 lessons in this course.