Social Skills

Abstract

Social skills, which MeSH classifies under social behavior, are the learned, observable behaviours that let a person interact competently with others — initiating conversation, reading social cues, cooperating, and resolving conflict. This article distinguishes social skills from the broader social competence that judges them and from the social cognition that underlies them, and develops the field's central claim: that these are acquired behaviours, not fixed traits, so deficits can be classified and taught. It traces the social-information-processing model from cue to action, the assessment instruments built on it, the links between skill deficits and peer rejection, loneliness, and depression, and the evidence that structured training works. Three interactive demonstrations let readers classify a skill deficit, step a social cue through the processing loop, and convert a training effect size into a percentile gain.

Keywords: social skills, social competence, social information processing, social skills training, assessment

Almost everything a person wants from other people is obtained through behaviour: a greeting, a question, a joke, an apology, a well-timed silence. Social skills are the name for the specific, learnable actions that make those exchanges go well, and for the capacity to deploy them at the right moment and in the right measure. They are not charm, not a personality type, and not an attitude. They are performances — things a person does, observable and countable, that can be absent, present, clumsy, or fluent, and that, being learned, can be taught to someone who lacks them.

That last fact is what makes social skills a topic for cognitive and behavioural science rather than for folk wisdom. If competent social behaviour were a fixed trait, there would be little to study and nothing to train. Because it is instead a repertoire of acquired behaviours governed by how a person reads and responds to social situations, it can be analysed into components, measured against a standard, and improved by instruction — and its failure can be diagnosed with the same precision. This article develops that analysis, beginning with the definitions the surrounding vocabulary routinely blurs.

What Social Skills Are

Social skills are the discrete, learned behaviours a person uses to interact effectively and appropriately with others. McFall's influential reformulation drew the line that organizes the whole field: social skills are the specific behaviours a person exhibits, whereas social competence is an evaluative judgement — made by some observer, against some criterion — that those behaviours were adequate to the task (McFall, 1982). The distinction is not pedantic. Skills are what a person does; competence is a verdict on the result. A person can perform every skill in the manual and still be judged incompetent in a situation that called for something else, and two observers can judge the same behaviour differently. Keeping the behaviour separate from the verdict on it is the first discipline the topic demands, because interventions act on behaviour while outcomes are measured as competence.

Two further constructs must be held apart from social skills themselves. Social competence sits above the skills as the summary judgement of their effectiveness; Rose-Krasnor's analysis treats competence as effectiveness in interaction, a transactional and context-bound quality that cannot be read off any single behaviour (Rose-Krasnor, 1997). Social cognition sits below the skills as the mental processing — perceiving, interpreting, and deciding — that generates them; it is the machinery, not the output. Social skills are the observable middle layer: the behaviours through which cognition becomes competence or fails to.

Social Skills as Learned Behavior

The organizing premise of the field is that social skills are acquired, not innate — and the practical payoff of that premise is a taxonomy of failure. Gresham's behavioural account defines social skills as socially acceptable learned behaviours that enable effective interaction and avoid socially unacceptable responses, and, crucially, distinguishes kinds of deficit by their cause (Gresham et al., 2001). An acquisition deficit means the person never learned the skill or cannot discriminate when to use it — a can't-do problem. A performance deficit means the person has the skill but does not perform it at acceptable levels — a won't-do problem, often because competing problem behaviour is more reinforcing, or because the skill is not prompted or rewarded in the setting. A fluency deficit means the skill is present but awkward for want of practice. The three call for different responses: acquisition deficits need direct instruction, performance deficits need antecedent prompts and changed contingencies, and fluency deficits need rehearsal. Mistaking one for another is how training fails — teaching a skill the child already has, or prompting a skill the child has never learned.

That skills are learned at all is the legacy of social learning theory. Bandura showed that much social behaviour is acquired by observing and imitating models, without direct reinforcement of the learner, and that whether an observed behaviour is then performed depends on expected consequences — a distinction that maps precisely onto the acquisition–performance split (Bandura, 1977). Modelling is, for the same reason, the first ingredient of nearly every training programme: if skills are learned by watching, then showing is teaching.

From Cue to Action: Social Information Processing

If skills are the observable behaviours, social information processing is the cognitive route that produces them. Argyle and Kendon framed social interaction as a skill in the motor sense — a continuous cycle of perception, translation into action, and correction from feedback, exactly like steering a car or playing tennis — and this cybernetic picture put the processing steps, not just the outputs, at the centre of analysis (Argyle & Kendon, 1967).

The dominant developmental formalization is Crick and Dodge's social-information-processing (SIP) model, which specifies the steps a child moves through in responding to a social situation: encoding cues, interpreting them, clarifying a goal, accessing possible responses, deciding among them, and enacting the choice (Crick & Dodge, 1994). Competent and maladaptive behaviour diverge at identifiable steps. The best-studied divergence is a hostile attribution bias at the interpretation step: a child who reads ambiguous conduct — a bump in the hallway — as deliberate hostility selects an aggressive response downstream, not because the response repertoire is impoverished but because the situation was misread.

The processing account also explains why social skill is so developmentally staged and so vulnerable to injury. Beauchamp and Anderson's SOCIAL model integrates the attention, communication, socio-emotional, and executive functions that underwrite social skill with the brain networks that mature to support them, showing why skills emerge on a developmental timetable and why acquired brain injury, autism, and attentional disorders disrupt them in characteristic ways (Beauchamp & Anderson, 2010). Skill, on this view, is the visible tip of a stack of maturing cognitive systems.

Assessing Social Skills

Because skills are behaviours, they can be measured — and the measurement problem is where the acquisition–performance distinction earns its keep. Riggio's Social Skills Inventory operationalized skill as a profile across basic dimensions of emotional and social expressivity, sensitivity, and control, giving a self-report map of a person's communicative repertoire rather than a single global score (Riggio, 1986). For children, the most widely used instruments are the rating scales developed by Gresham and Elliott — the Social Skills Rating System and its successor, the Social Skills Improvement System — which gather ratings from teachers, parents, and the child across domains such as cooperation, assertion, responsibility, empathy, and self-control, and pair them with a problem-behaviour scale so that a performance deficit driven by competing behaviour can be distinguished from a genuine absence of skill (Gresham, 2016). Rose-Krasnor's analysis supplies the caution that no instrument escapes: because competence is a context-bound judgement, any score is an abstraction from situations, and a skill adequate in one setting may be inadequate in another (Rose-Krasnor, 1997).

Deficits and Their Consequences

Social skill deficits are not a cosmetic problem; they forecast serious outcomes, because the human need to belong makes competent interaction a gateway to nearly everything else. Baumeister and Leary's need-to-belong hypothesis holds that the drive to form and maintain lasting, positive relationships is a fundamental human motivation, so the behaviours that secure relationships are consequential out of proportion to their apparent triviality (Baumeister & Leary, 1995). In childhood, the consequences show up first in the peer group. Parker and Asher's review established that children who are poorly accepted by peers — many of them for want of the skills that make a child a desirable playmate — are at elevated risk for later difficulties including dropping out and externalizing problems, making peer rejection one of the earliest and most durable predictors in developmental psychopathology (Parker & Asher, 1987).

Across the lifespan the clearest link is with mood. Segrin's work established a reciprocal relationship between social skills deficits and depression: poor interpersonal skill generates the rejection, conflict, and loneliness that precipitate and maintain depression, while depression in turn degrades the skill, in a cycle each side feeds (Segrin, 2000). The bidirectionality matters for intervention — treating the mood without rebuilding the skill, or the reverse, leaves half the cycle intact — and it is the adult mirror of the childhood peer-rejection story: in both, a behavioural deficit and a social-emotional cost sustain each other.

Training and Intervention

The strongest practical claim the field makes is that social skills can be taught, and the evidence supports it when the training is done well. Spence's account of social skills training with children and young people lays out the core method — identify the target skills, model them, coach and rehearse them through role-play, give corrective feedback, and plan explicitly for generalization to real settings — and reviews the evidence for its effectiveness across clinical and developmental populations (Spence, 2003). The generalization step is the one most often skipped and most often decisive: a skill rehearsed in a clinic that is never cued in the classroom or playground is a performance deficit waiting to happen, which is why Gresham and colleagues attribute many disappointing training results not to the method but to too little dosage and too little programming for transfer (Gresham et al., 2001).

At scale, the universal version of this work is social and emotional learning (SEL) in schools. Durlak and colleagues' meta-analysis of 213 school-based universal SEL programmes found significant improvements in social and emotional skills, attitudes, behaviour, and — notably — an 11-percentile-point gain in academic achievement, establishing that teaching these skills to all children, not only those with deficits, pays measurable dividends (Durlak et al., 2011). A follow-up synthesis showed the benefits persist: gains in skills and well-being, and reductions in conduct problems and distress, were detectable months to years after the programmes ended (Taylor et al., 2017). For children with autism spectrum disorder, for whom skill deficits are a core feature, group social skills interventions produce reliable if modest gains: a meta-analysis found consistent benefits on social competence and peer relations (Gates et al., 2017), and a randomized controlled trial confirmed improvements that were clearer on parent and clinician ratings than on blinded measures, underscoring how much the outcome depends on who is judging and where (Choque Olsson et al., 2017).

Discussion

The study of social skills is unusual among social-science topics in how directly its theory licenses its practice. The founding move — McFall's separation of skills from the competence judgement, and Gresham's split of acquisition from performance deficits — is not merely conceptual tidiness; it tells a clinician which of three different things to do when a child does not behave well with peers. The social-information-processing model does the same for cognition, locating the point in the sequence where behaviour went wrong so that intervention can be aimed there rather than at the behaviour's surface. A field whose central distinctions map onto distinct interventions is a field whose theory is doing work.

The through-line is that social skill is a behaviour, not a trait, and therefore both diagnosable and teachable. That premise carries the field's optimism — deficits can be remediated — and also its hardest problem: generalization. Skills reliably improve within training; whether they transfer to the settings and relationships that matter is the question on which real-world benefit turns, and the one the evidence is most equivocal about. The embedded, whole-population SEL programmes do best precisely because they dissolve the transfer problem by training skills where they are used.

The limitations track the measurement problem. Competence is a context-bound judgement, so every instrument abstracts away the situation; ratings disagree across informants and settings; and the training literature's reliance on proximal, often unblinded outcomes — parents and teachers who know a child received an intervention — inflates apparent effects, as the autism RCT's split between rated and blinded measures makes plain. The behaviours are real and countable, but the verdict on them never fully is.

Current Directions

The clearest current trend is the move from pulled-out training to embedded, universal provision, and from short-term gains to durable ones. Taylor and colleagues' demonstration that SEL benefits persist months to years after programmes end reframed the question from does it work now to does it last, and shifted attention to the programme features — dosage, generalization programming, developmental timing — that predict persistence (Taylor et al., 2017). A second strand is the sharpening of intervention for specific populations: the autism social skills literature has matured from asking whether group training helps to specifying for whom, by what mechanism, and measured how, with the recurring finding that effects are robust on proximal ratings and thinner on blinded or generalized measures (Gates et al., 2017)(Choque Olsson et al., 2017). The third is the continued integration of the behavioural and cognitive traditions — pairing the observable-skill taxonomy with the processing model and its neurodevelopmental substrate — so that assessment can target not only which skill is absent but which processing step produced its absence (Beauchamp & Anderson, 2010)(Gresham, 2016). The open questions across all three are the same: how to measure skill in context rather than in abstraction, how to secure transfer, and how much of the long-run social and clinical benefit the trainable behaviours actually carry.

Common Misconceptions

Social skills are a personality trait a person either has or lacks.

They are learned behaviours, not a fixed disposition. Because they are acquired by instruction, modelling, and practice, they can be taught to someone who lacks them and sharpened in someone who has them — which is the entire basis of social skills training (Gresham et al., 2001).

Social skills and social competence are the same thing.

They are different. Social skills are the specific behaviours a person performs; social competence is an observer's judgement that those behaviours were effective and appropriate. A person can perform the behaviours and still be judged incompetent in a situation that called for something else (McFall, 1982).

If a child does not use a skill, they must not have it.

Not necessarily. A performance deficit — having the skill but not using it, often because competing behaviour pays off better or the setting never prompts it — looks identical to an acquisition deficit from the outside but calls for a completely different intervention (Gresham et al., 2001).

Aggressive children simply lack social skills.

Often the problem is upstream of the skills. In the social-information-processing model, many aggressive children misread ambiguous situations as hostile at the interpretation step, so an otherwise available aggressive response is selected — a cognitive bias, not a missing behaviour (Crick & Dodge, 1994).

Demo 1 — Classifying a Skill Deficit

Gresham argues that a social skill can fail in three different ways, and each calls for a different intervention. Answer the three diagnostic questions and the demo returns the classification and the matched remedy. Note that the second and third questions matter only once the skill is in the repertoire.

Demo 2 — The Social-Information-Processing Loop

Crick and Dodge model the steps between a social cue and a response. Step through an ambiguous provocation — a peer bumps into you and your books fall — and at the interpretation step choose how you read it. A hostile reading cascades through every later step toward an aggressive response, even though the aggressive act is available either way.

Demo 3 — Reading an Effect Size

A training study reports its result as Cohen's d, the gap between a trained group and a control in standard-deviation units. Drag d to shift the trained distribution (navy) away from the control (grey). The percentile is the normal cumulative probability Φ(d): where the median trained participant now sits in the untrained distribution — Cohen's U₃ overlap.

At d = 0.57, the median trained participant out-skills about 72% of untrained peers (the 50th percentile with no training). A d well under one standard deviation, easy to dismiss as small, still moves a typical child past a fifth to a quarter of their peers.

Worked Example

Training studies report their results as effect sizes, and an effect size is easy to state but hard to feel. The standardized mean difference (Cohen's d) can be turned into something concrete — a percentile standing — using the normal distribution, and doing so makes the magnitude of a social skills programme legible.

Durlak and colleagues' meta-analysis reported a mean effect of about d = 0.57 for the social and emotional skills targeted by school-based SEL programmes (Durlak et al., 2011). Cohen's d expresses the gap between a trained group and an untrained control in standard-deviation units:

d = (M_trained − M_control) ÷ SD_pooled = 0.57

To read that as a percentile, take a child who would have sat exactly at the median of the untrained distribution — the 50th percentile — and ask where the d-sized shift lands them in that distribution. The answer is the normal cumulative probability at d:

percentile = Φ(0.57) ≈ 0.716, or the 72nd percentile

So the average participant moves from the 50th percentile to roughly the 72nd — a gain of about 22 percentile points in measured social-emotional skill. Equivalently, Cohen's U₃ says that the median trained child now out-skills about 72% of untrained peers, up from 50%. The same conversion applied to the programmes' d ≈ 0.27 effect on academic achievement gives Φ(0.27) ≈ 0.606 — the 61st percentile, the 11-percentile-point achievement gain the authors headline. The lesson of the arithmetic is that a d under one standard deviation, easy to dismiss as small, moves a typical child past a fifth to a quarter of their peers. The third interactive demonstration lets the reader set d and read off both the percentile standing and the U₃ overlap directly.

Key Researchers

Michael Argyle

(University of Oxford; 1925–2002). Launched the experimental analysis of social performance, treating social interaction as a trainable motor skill governed by perception, action, and feedback rather than fixed personality. Wikipedia · Wikidata

Nicki R. Crick

(University of Minnesota; 1958–2012). Co-developed the social-information-processing model of children's social adjustment and identified relational aggression as a distinct form of skill deficit. Wikipedia · Wikidata

Kenneth A. Dodge

(Duke University). Co-authored the social-information-processing framework linking the encoding and interpretation of social cues to competent versus aggressive behaviour, the dominant cognitive account of children's social skill. Faculty page · Google Scholar · Wikipedia

Stephen N. Elliott

(Arizona State University). Co-developed the Social Skills Rating System and its successor the Social Skills Improvement System, the most widely used standardized measures of children's social skills. ORCID · Faculty page · Google Scholar

Frank M. Gresham

(Louisiana State University; 1949–2026). Defined social skills as acquired, learnable behaviours and distinguished acquisition from performance deficits, the conceptual basis of modern social skills training and assessment. Faculty page · Google Scholar

Chris Segrin

(University of Arizona). Established the reciprocal link between social skills deficits and psychopathology, especially depression and loneliness, showing poor interpersonal competence as both cause and consequence of distress. ORCID · Faculty page · Google Scholar

Susan H. Spence

(Griffith University). Built evidence-based social skills training protocols for children and young people and clarified the theory and practice of social competence intervention in clinical and school settings. ORCID · Faculty page · Google Scholar · Wikipedia

Glossary

Acquisition deficit.

A skill failure in which the person never learned the skill or cannot tell when to use it — a can't-do problem that calls for direct instruction.

Cohen's d.

The standardized mean difference between two groups, expressed in pooled standard-deviation units; the common currency of effect size in training research.

Fluency deficit.

A skill failure in which the behaviour is present but awkward or slow for want of practice; remediated by rehearsal rather than instruction.

Hostile attribution bias.

The tendency to interpret ambiguous social conduct as deliberately hostile; a divergence at the interpretation step of the social-information-processing model linked to aggression.

Need to belong.

Baumeister and Leary's hypothesis that the drive for lasting, positive relationships is a fundamental human motivation, making the behaviours that secure relationships consequential.

Performance deficit.

A skill failure in which the person has the skill but does not use it — a won't-do problem, often from competing behaviour or an unprompting setting; remediated by changed contingencies.

Relational aggression.

Harming others through manipulation of relationships and social standing rather than physical force; identified by Crick as a distinct behavioural pattern.

Social and emotional learning (SEL).

School-based, usually universal instruction in social and emotional skills; the whole-population form of social skills training.

Social behavior.

Any behaviour directed toward or influenced by other members of the same species; the superordinate category under which MeSH files social skills.

Social cognition.

The mental processing — perceiving, interpreting, and deciding about social information — that generates social behaviour; the machinery beneath the skills.

Social competence.

An observer's judgement, against some criterion, that a person's social behaviour was effective and appropriate; a verdict on the skills, distinct from the skills themselves.

Social information processing.

Crick and Dodge's model of the steps between a social cue and a response: encoding, interpretation, goal clarification, response access, response decision, and enactment.

Social skills

The specific, learned, observable behaviours a person uses to interact effectively and appropriately with others.

Social skills training (SST).

Structured instruction in social skills through modelling, coaching, rehearsal, feedback, and planned generalization to real settings.

U₃.

Cohen's overlap statistic: the percentage of the untrained distribution that the median trained participant exceeds, read directly from the normal CDF of d.

Frequently Asked Questions

What are social skills?

Social skills are the specific, learned, observable behaviours a person uses to interact effectively and appropriately with others: starting a conversation, reading social cues, cooperating, asserting a need, and resolving conflict. They are things a person does, which is what makes them countable, teachable, and distinct from traits or attitudes.

How are social skills different from social competence?

Social skills are the behaviours themselves; social competence is a judgement, made by some observer against some standard, that those behaviours were effective and appropriate. A person can perform a skill flawlessly and still be judged incompetent in a situation that needed something different, so the behaviour and the verdict on it have to be kept separate.

Can social skills be taught?

Yes. Because they are learned rather than innate, they respond to structured training: modelling the skill, coaching and rehearsing it through role-play, giving corrective feedback, and planning for its use in real settings. Meta-analyses of school-based programmes find measurable gains in skills, behaviour, and even academic achievement.

Why does a child have a skill but not use it?

That is a performance deficit, as opposed to an acquisition deficit. The child has learned the skill but does not perform it, often because competing problem behaviour is more rewarding, or because the setting never prompts or reinforces the skill. It looks identical to never having learned the skill but needs a different intervention: changed prompts and consequences, not instruction.

What is the social-information-processing model?

It is Crick and Dodge's account of the mental steps between a social situation and a response: encoding the cues, interpreting them, clarifying a goal, accessing possible responses, deciding among them, and enacting the choice. It lets researchers locate where competent and maladaptive behaviour diverge: for instance, a hostile attribution bias at the interpretation step.

Are social skill problems linked to mental health?

Strongly, and in both directions. Poor social skills generate the rejection, conflict, and loneliness that help precipitate and maintain depression, while depression in turn degrades social skill, in a reciprocal cycle. In childhood, poor peer acceptance, which often reflects skill deficits, is one of the most durable predictors of later difficulty.

Do social skills interventions work for autistic people?

Group social skills interventions produce reliable if modest gains for people with autism spectrum disorder, for whom skill deficits are a core feature. Effects are clearest on parent- and clinician-rated measures and thinner on blinded or generalized ones, so the benefit is real but its size depends heavily on who is judging and in what setting.

What is the hardest part of social skills training?

Generalization: getting a skill that improves during training to transfer to the real relationships and settings that matter. A skill rehearsed in a clinic but never cued at school or home stays unused, which is why the most effective programmes embed training in everyday settings rather than pulling learners out for it.

References

Argyle, M., & Kendon, A. (1967). The experimental analysis of social performance. Advances in Experimental Social Psychology, 3, 55–98. https://doi.org/10.1016/S0065-2601(08)60342-1

Bandura, A. (1977). Social learning theory. Prentice-Hall.

Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117(3), 497–529. https://doi.org/10.1037/0033-2909.117.3.497

Beauchamp, M. H., & Anderson, V. (2010). SOCIAL: An integrative framework for the development of social skills. Psychological Bulletin, 136(1), 39–64. https://doi.org/10.1037/a0017768

Choque Olsson, N., Flygare, O., Coco, C., Görling, A., Råde, A., Chen, Q., Lindstedt, K., Berggren, S., Serlachius, E., Jonsson, U., Tammimies, K., Kjellin, L., & Bölte, S. (2017). Social skills training for children and adolescents with autism spectrum disorder: A randomized controlled trial. Journal of the American Academy of Child & Adolescent Psychiatry, 56(7), 585–592. https://doi.org/10.1016/j.jaac.2017.05.001

Crick, N. R., & Dodge, K. A. (1994). A review and reformulation of social information-processing mechanisms in children's social adjustment. Psychological Bulletin, 115(1), 74–101. https://doi.org/10.1037/0033-2909.115.1.74

Durlak, J. A., Weissberg, R. P., Dymnicki, A. B., Taylor, R. D., & Schellinger, K. B. (2011). The impact of enhancing students' social and emotional learning: A meta-analysis of school-based universal interventions. Child Development, 82(1), 405–432. https://doi.org/10.1111/j.1467-8624.2010.01564.x

Gates, J. A., Kang, E., & Lerner, M. D. (2017). Efficacy of group social skills interventions for youth with autism spectrum disorder: A systematic review and meta-analysis. Clinical Psychology Review, 52, 164–181. https://doi.org/10.1016/j.cpr.2017.01.006

Gresham, F. M. (2016). Social skills assessment and intervention for children and youth. Cambridge Journal of Education, 46(3), 319–332. https://doi.org/10.1080/0305764X.2016.1195788

Gresham, F. M., Sugai, G., & Horner, R. H. (2001). Interpreting outcomes of social skills training for students with high-incidence disabilities. Exceptional Children, 67(3), 331–344. https://doi.org/10.1177/001440290106700303

McFall, R. M. (1982). A review and reformulation of the concept of social skills. Behavioral Assessment, 4(1), 1–33.

Parker, J. G., & Asher, S. R. (1987). Peer relations and later personal adjustment: Are low-accepted children at risk? Psychological Bulletin, 102(3), 357–389. https://doi.org/10.1037/0033-2909.102.3.357

Riggio, R. E. (1986). Assessment of basic social skills. Journal of Personality and Social Psychology, 51(3), 649–660. https://doi.org/10.1037/0022-3514.51.3.649

Rose-Krasnor, L. (1997). The nature of social competence: A theoretical review. Social Development, 6(1), 111–135. https://doi.org/10.1111/j.1467-9507.1997.tb00097.x

Segrin, C. (2000). Social skills deficits associated with depression. Clinical Psychology Review, 20(3), 379–403. https://doi.org/10.1016/S0272-7358(98)00104-4

Spence, S. H. (2003). Social skills training with children and young people: Theory, evidence and practice. Child and Adolescent Mental Health, 8(2), 84–96. https://doi.org/10.1111/1475-3588.00051

Taylor, R. D., Oberle, E., Durlak, J. A., & Weissberg, R. P. (2017). Promoting positive youth development through school-based social and emotional learning interventions: A meta-analysis of follow-up effects. Child Development, 88(4), 1156–1171. https://doi.org/10.1111/cdev.12864