Understanding When Parental Praise Leads to Optimal Child Outcomes

Abstract

The current work examined conditions under which parental praise leads to higher academic achievement and better psychological health in schoolchildren. We tested the hypothesis that perceptions of accurate praise, both by parents and by children, are associated with outcomes optimal for children. Our results showed that parents’ perceptions of over- or underpraising (vs. accurately praising) their children’s schoolwork predicted poorer school performance and higher depression in children. From children’s perspectives, perceived under- and overpraise by parents predicted poorer school performance and higher depression. However, when children felt that their parents’ praise was slightly (but not majorly) overstated, this had at least as beneficial effects as when they felt the praise accurately reflected reality. For parents and educators, these results underline the importance of basing praise of children on actual performance and the need to pay careful attention to how praise is perceived by the child.

Keywords

academic performance depression parental praise perceived accuracy

Praise is a tool commonly used by parents to motivate their children and bolster their self-confidence. Potentially influenced by self-esteem and positive-thinking movements, many parents hold the belief that praise is beneficial to their children’s academic achievement and psychological well-being (Mueller & Dweck, 1996; Reid & Valsiner, 1986). However, evidence for the effectiveness of praise is mixed. Some data support the lay idea that praise helps children become more motivated (Cameron & Pierce, 1994; Taylor, Fisher, & Ilgen, 1984), socially skilled (Garner, 2006; Hastings, McShane, Parker, & Ladha, 2007), and self-confident and well-adjusted (Delin & Baumeister, 1994; Weiner, 1992). However, detrimental consequences of praise have also been reported, including its capacity to undermine children’s intrinsic motivation (Kohn, 1993; Lepper & Greene, 1978; Weiner, 1992), impair task performance (Baumeister, Hutton, & Cairns, 1990; Grusec, 1991; Mueller & Dweck, 1998), and discourage challenge seeking, particularly among children with low self-esteem (Brummelman, Thomaes, de Castro, Overbeek, & Bushman, 2014).

In attempts to reconcile these contradictory findings, researchers have tried to identify conditions under which praise helps rather than hurts children’s academic performance and psychological well-being. In particular, recommendations have been offered regarding what should and should not be the focus of praise and how praise should be delivered. Different lines of research have looked at offering praise based on person versus process (Gunderson et al., 2013; Kamins & Dweck, 1999; Pomerantz & Kempner, 2013) or intelligence versus effort (Mueller & Dweck, 1998) and in delivering praise in generic versus nongeneric terms (Cimpian, Arce, Markman, & Dweck, 2007) or in an informational versus a controlling manner (Kast & Connor, 1988).

However, the accuracy of praise, a factor playing a critical role in the effectiveness of performance feedback (Kim, Chiu, & Zou, 2010; Kinicki, Prussia, Wu, & McKee-Ryan, 2004), has been the focus of research less frequently (Henderlong & Lepper, 2002). A few recent studies have only indirectly tapped on the effectiveness of unwarranted and inflated praise on children’s academic performance and provided suggestive evidence of adverse effects. Brummelman, Thomaes, de Castro, Overbeek, and Bushman (2014) found that children with low self-esteem were less likely to take on challenges upon receiving inflated praise. Similarly, Murayama, Pekrun, Suzuki, Marsh, and Lichtenfeld (2016) showed that unrealistically high parental aspirations for their children’s academic achievement, called “parental overaspiration,” was associated with poorer academic performance over time.

In the present study, we addressed this gap in knowledge by directly examining how the accuracy of praise, as perceived by both parents who give it and children who receive it, predicts academic performance and psychological well-being among children.

Children’s Perceptions of the Accuracy of Praise Received

We hypothesized that praise, a form of performance feedback, has detrimental effects on children’s achievement and psychological well-being, when it is perceived to overstate or understate reality, instead of reflecting it accurately. Prior research provides suggestive evidence that when there is a discrepancy between people’s subjective assessment of performance and the feedback they receive, people respond to this discrepancy with increased emotional distress and self-destructive behaviors. For instance, when people receive overly positive feedback, they experience heightened concerns about failure to meet high expectancies, which in turn leads them to avoid further learning experience (Brummelman et al., 2014), becoming more vulnerable to emotional distress in the process (Kim & Chiu, 2011). Under increased pressure to sustain their inflated self-views, they also engage in self-handicapping strategies by putting forth less preparatory effort for upcoming tasks. In contrast, when high performers receive overly negative feedback, their self-views become more critical and their susceptibility to distress increases (Kim & Chiu, 2011). Starting to doubt their abilities, they end up withholding their preparatory effort for upcoming tasks (Kim, Chiu, & Zou, 2010). These findings suggest that feedback incongruent with actual performance is detrimental to both performance and emotional well-being of the recipient.

Against this background, we predicted that praise perceived to be based on actual performance yields the most desirable outcomes for children. In other words, parental praise perceived to be accurate should be more effective in motivating children to adopt or sustain strategies that lead to better academic performance. Accordingly, we expected that parental praise perceived to be accurate would be associated with better psychological health and higher school success.

The Parental Perception of the Accuracy of Praise Given

We also examined how parental perceptions of their praise related to outcomes in their children. We predicted that negative outcomes may result when parents perceive their own praise as inaccurate. Indeed, the gap that a parent feels between reality and their actions could affect children negatively if their child is aware of the gap. It also stands to reason that parents who admit to over- or underpraising their children may be less effective parents or have more difficult relationships with their children in general, which may be another factor predicting poor child outcomes. Thus, we hypothesized that parental perceptions of accurate praise would be associated with higher school performance and psychological well-being in children.

Overview of the Study

In a sample of South Korean elementary school students and their parents, we examined how parental praise for schoolwork related to academic achievement and psychological well-being among children. Specifically, we assessed how much parents thought they over- or underpraised their children for schoolwork, and how much the children perceived that their parents over- or underpraised them. Our stated hypothesis was that the best outcomes for academic achievement and psychological well-being should be observed when praise was perceived by both parents and children to reflect performance accurately. If our hypothesis had merit, there should be a quadratic relationship between perceptions of praise and the children’s performance and psychological well-being. That is, when praise is perceived as either overstated or understated, the extent of that overstatement or understatement would be negatively associated with academic achievement and psychological well-being. By contrast, however, accurate praise would be associated with favorable academic achievement and psychological well-being. The same patterns should hold regardless of whether the perceptions belong to the parent or the child—with the caveat that children’s perceptions of a slight overstatement could predict outcomes comparable to accurate praise.

Method

Participants

Preliminary power analyses revealed that approximately 300 participants were required for the study to detect a likely effect size of r = 0.15 with at least an 80% power. All third-, fourth-, and fifth-grade children from a private elementary school (337 students; 161 males and 176 females) in Seoul, South Korea, together with one parent per child, participated in the study.¹ The student sample comprised 118 third graders (8–9 years), 99 fourth graders (9–10 years), and 120 fifth graders (10–11 years). We also recruited 298 parents (mean age, 44.00 ± 3.52 years); of these, 88.59% were mothers and 11.41% were fathers. The school sent a notice to parents and sought their permission to administer the survey about two weeks before both the students and parents were asked to complete the questionnaire.

Student Questionnaire

Upon completing a written informed consent form, students completed their questionnaires during regular class time. Along with several filler questions, they were asked to indicate the extent to which the praise their parents give them on their academic performance (e.g., homework, exam scores, report cards) was overstated or understated, as compared with their perception of actual school performance. We specifically asked: “When your mother (or father) praises you, to what extent does she (or he) overpraise or underpraise you for your academic performance, given your actual performance?” with responses provided on a 7-point Likert-type scale ranging from 1 (very much understated) to 7 (very much overstated), and the scale midpoint labeled “accurately reflecting my school performance.” The students made separate judgments for the praise given by their father and mother. They also completed a Beck Depression Inventory-II (Beck, Steer, & Brown, 1996), a widely used 21-item scale to measure the severity of depression in adolescents and adults. The original scale was slightly modified to be comprehensible for Korean children, and 1 item (regarding loss of interest in sex) was dropped. Students indicated the extent to which they experienced different depression symptoms (e.g., sadness, self-dislike, and crying), with each item measuring the severity of that symptom on a scale ranging from 0 (least severity) to 3 (most severity). Thus, the total score could range between 0 and 60.

Parent Questionnaire

On the day students completed their questionnaire, they were asked to take the parent questionnaire home. Upon completing a written informed consent form, the parents then completed the parent questionnaire. In the questionnaire, they were asked to indicate the extent to which the praise they gave their children for schoolwork was understated or overstated, as compared to their children’s actual school performance. We specifically asked: “When you praise your children, to what extent do you over- or underpraise their academic performance given their actual performance?” with responses again given on a 7-point Likert-type scale ranging from 1 (very much understated) to 7 (very much overstated), and the scale midpoint labeled “accurately reflecting my child’s school performance.” Parents were also asked to indicate (a) how much they praised their children for schoolwork, and (b) how much they monitored their children’s academic performance, again using a 7-point Likert-type scale, but ranging from 1 (not at all) to 7 (a lot). We aimed to control for these variables to examine the unique effect of parental perception of praise accuracy on each dependent variable. To control for the effects of parental education and family income on the dependent variables further, we also asked parents to indicate their highest academic qualification and to give their monthly household income. Completed questionnaires were returned to the school by students.

Grade Point Average (GPA)

At the end of the semester, we obtained students’ semester performance scores for three major subjects—language arts, mathematics, and English²—from the school registrar. Scores in each subject could range from 0 to 100. In the present study, we averaged the performance scores for the three subjects (reliability coefficient α = .82) to create a composite index of students’ academic achievement, the GPA. We also obtained students’ GPA for the previous semester from the average of the same three subjects (reliability coefficient α = .83), with the purpose of treating it as a covariate in the data analysis.

Results

Descriptive statistics and zero-order correlations among variables are presented in Table 1.

Table 1.

Descriptive Statistics and Correlations.

	Variable	Mean	SD	1	2	3	4	5	6	7	8	9	10	11	12
1.	PAP	4.67	1.18	—
2.	PAC	4.28	0.68	.20**	.45
3.	MPAC	4.28	0.89	.16**	.82**	—
4.	PPAC	4.28	0.82	.16**	.79**	.29**	—
5.	Previous GPA	89.09	9.00	−.14*	−.01	.00	−.06	.83
6.	GPA	88.81	9.47	−.13*	.02	.01	−.01	.93**	.82
7.	Depression	7.08	7.57	.05	−.08	.01	−.10	−.25**	−.30**	.82
8.	Amount of praise	5.11	1.08	.51**	.12*	.12*	.07	.07	.05	−.07	—
9.	Parent’s monitoring	4.70	1.11	.05	.01	.04	−.02	.16**	.15*	−.02	.36**	—
10.	Mother’s education level	3.05	0.61	.05	.04	.08	.00	.11	.13*	.07	.02	−.01	—
11.	Father’s education level	3.22	0.68	.09	.06	.09	.01	.04	.03	.07	−.01	−.08	.43**	—
12.	Household income	4.52	0.72	.05	−.02	−.03	−.02	.08	.06	−.07	.14*	.04	.15**	.13*	—

Note. n = 294–326, PAP = praise accuracy perceived by parent; PAC = praise accuracy perceived by child; MPAC = maternal praise accuracy perceived by child; PPAC = paternal praise accuracy perceived by child. The coefficients on the diagonal in bold are Cronbach’s α of each variable.

*p < .05. **p < .01.

Parental Perception of the Accuracy of Praise Given

The average perceived accuracy of parental praise was 4.67 ± 1.18, which was reliably different from 4.00, the midpoint of the scale; t(297) = 9.84, p = .001. This suggested that parents generally saw themselves as giving their children higher praise than their school performance warranted. The children’s average GPA and depression levels were 88.81 ± 9.47 and 7.08 ± 7.57, respectively.

To test our hypotheses, we fitted a polynomial regression equation to each dependent variable (GPA and depression level) with the linear and quadratic effects of parents’ perceived accuracy of praise as predictors (each dependent variable = a + bX + cX², where X represents parents’ perceived accuracy of praise and a, b, c were empirically estimated parameters). Note that the perceived accuracy of parental praise was centered using 4 (the scale midpoint), and the quadratic term was created with the centered predictor prior to analyses, so that the linear effect would represent the simple linear effect of the perceived praise accuracy when the praise is perceived to be accurate. For the GPA, the linear effect was not significant (B = −0.35, t(285) = −0.63, p = .53), but the predicted quadratic effect was significant (B = −0.64, t(285) = −1.98, p = .04). To decompose the quadratic effect, we tested the simple linear effects of perceived accuracy when the parents perceived the praise to be overstated (at a value of 6 on the 7-point scale) or understated (at a value of 2 on the 7-point scale; Judd, McClelland, & Ryan, 2011). We found that, when the parents perceived their praise to be overstated, the simple main effect was negatively significant (B = −2.93, t(285) = −2.74, p = .01). But, although still in the predicted direction, the simple linear effect of the perceived accuracy of the praise on GPA was not significant when the parents perceived their praise to be understated (B = 2.22, t(285) = 1.31, p = .19). As evident in Figure 1, perceptions of both over- and understated praise by parents were associated with lower GPAs for their children. By contrast, perceptions of accurate praise were associated with the highest GPA levels.

Figure 1.

The effect of parents’ perceived praise accuracy on children’s grade point averages (left panel) and depression levels (right panel). Note. Parents’ perceived praise accuracy was plotted along the x-axis from 1 (very much understated) to 7 (very much overstated). The midpoint (4) was labeled “accurately reflecting my child’s school performance.”

The results were comparable for depression. The quadratic (B = 0.76, t(284) = 2.83, p = .01) effect was significant, while the linear effect was not (B = −0.46, t(284) = −1.01, p = .31). When the parents perceived their praise to be overstated (at the value of 6), the simple linear effect on depression was positively significant (B = 2.57, t(284) = 2.92, p = .004), but when parents perceived their praise to be understated (at the value of 2), the simple linear effect of the perceived accuracy of praise on depression was negatively significant (B = −3.50, t(284) = −2.50, p = .01). As seen in Figure 1, perceptions of both overstated and understated praise by parents were associated with higher levels of depression in their children, while perceived accurate praise effectively predicted the lowest depression levels.

Children’s Perception of the Accuracy of Praise Received

The average perceived accuracy of praise given by mothers (4.28 ± 0.87) and fathers (4.28 ± 0.82) were not significantly different. Also, the quadratic effect of the perceived praise from mothers (95% confidence interval [CIs] = [−2.33, −0.31] for GPA and [0.45, 2.00] for depression) and fathers (95% CIs [−2.52, −0.38] for GPA and [−0.04, 2.24] for depression) were not different. Therefore, we averaged the children’s perceptions of praise from both mothers and fathers to serve as the index of children’s perception of praise from their parents. Children’s average perception of praise received from parents (4.28 ± 0.68) was reliably different from 4.00, t(321) = 7.40, p = .001, suggesting that children perceived the praise given by their parents to be slightly overstated.

To test our hypothesis that children’s perceptions of both over- and understated praise would be associated with less desirable outcomes, we first fitted a polynomial regression equation to each dependent variable with the linear and quadratic effects of perceived praise accuracy. Again, the praise accuracy perceived by children was centered at the scale midpoint 4 prior to analyses. For GPA, both the linear (B = 2.15, t(315) = 1.98, p = .04) and quadratic (B = −1.33, t(315) = −2.40, p = .02) effects were significant. Significant results were also obtained regarding perceived praise accuracy on depression for both the linear (B = −2.47, t(319) = −2.80, p = .01) and quadratic (B = 1.15, t(319) = 2.54, p = .01) effects. When the children perceived their parent’s praise to be overstated, the simple linear effect was significant in the expected direction for the GPA (B = −3.18, t(315) = −1.96, p = .05) and was marginally significant for depression (B = 2.12, t(319) = 1.61, p = .10); also, when children perceived the praise to be understated, the simple linear effect was again significant in the expected direction for the GPA (B = 7.49, t(315) = 2.41, p = .02) and depression (B = −7.07, t(319) = −2.80, p = .01).

Figure 2 shows the relationship between children’s perception of the accuracy of the praise received from their parents and their GPA and depression scores. A visual examination of this graph reveals that, for both variables, the optimal outcomes were obtained when children perceive their parents’ praise to be somewhere between accurate and slightly overstated, corresponding to the area between 4 and 5 on the 7-point Likert-type scale. This finding suggests that children benefit from perceiving their parents’ praise as somewhat—but not vastly—more generous than warranted by their actual performance.

Figure 2.

The relationship between children’s perception of praise accuracy by parents on grade point average (left panel) and depression scores (right panel). Note. Students’ perceived praise accuracy was plotted along the x-axis from 1 (very much understated) to 7 (very much overstated). The midpoint (4) was labeled “accurately reflecting my school performance.”

Do Children’s Perceptions of Praise Accuracy Matter beyond Their Parents’ Perceptions?

The findings above suggest that both the parents’ and children’s perceptions of praise accuracy were associated with the GPA and depression outcomes. We should note that the correlation between children’s and the parents’ perceptions of praise accuracy was significant but weak (r = 0.18, p = .002). This suggests that, while there is some overlap between the perceptions of children and their parents, this overlap is quite limited.

This raises the question of whether children’s perceptions might predict their outcomes independent of their parents’ perceptions. In other words, might the effect of a child’s perception of accuracy still matter after controlling for the effect of parental perception? To answer this question, we fitted a polynomial regression equation to each dependent variable with the linear and quadratic effects of both the parents’ (either the mother or father, whichever parent completed the questionnaire) and the children’s perceived praise accuracy, and the interaction of the two linear effects. This revealed that the quadratic effect of perceived praise accuracy among children was still reliable for GPA (B = −1.01, t(277) = −2.72, p = .01) and depression (B = 1.14, t(280) = 3.69, p = .001; see Table 2). This result suggests that the children’s perceptions of praise accuracy impacted their outcomes above and beyond their parents’ perceptions, regardless of what the parents thought about the accuracy of their praise.

Table 2.

Regression Analysis Predicting the Grade Point Average or Depression Score by Praise Accuracy perceived from Parents and Children.

	Outcomes
	GPA					Depression
Variable	B	SE	t	p	Partial Correlation [95% CI]	B	SE	t	p	Partial Correlation [95% CI]
PAC	1.84	0.84	2.20	.03	.13 [.01, .24]	−2.20	0.70	−3.16	<.01	−.18 [−.30, −.07]
PAC²	−1.01	0.37	−2.72	.01	−.16 [−.27, −.04]	1.14	0.31	3.69	<.01	.22 [.10, .32]
PAP	−0.24	0.55	−0.43	.67	−.03 [−.14, .09]	−0.43	0.46	−0.93	.35	−.06 [−.17, .06]
PAP²	−0.55	0.32	−1.72	.09	−.10 [−.22, .01]	0.65	0.27	2.47	.01	.15 [.03, .26]
PAC × PAP	−0.71	0.55	−1.28	.20	−.08 [−.19, .04]	0.41	0.46	0.88	.38	.05 [−.06, .17]
R ²	.075					.092

Note. PAC = praise accuracy perceived by child; PAP = praise accuracy perceived by parent; CI = confidence interval.

The Relation Between Children’s Perceptions of Praise Accuracy and Academic Performance After Controlling for Depression

It would not be surprising if depression in children negatively affected school performance. Children with greater levels of depression, for instance, could be particularly prone to perform poorly in school. Therefore, we also wanted to ascertain that the effect of children’s perceptions of praise accuracy on their GPA would hold even after controlling for their depression, which might be a determining factor in school success. Controlling for depression, we found that the quadratic (B = −0.98, t(314) = −1.80, p = .07) effect of the children’s perceptions on accuracy of their parents’ praise was still reliable on their GPA, but the linear effect (B = 1.38, t(314) = 1.30, p = .20) was not. In this model, as expected, the effect of depression on GPA was reliable (B = −0.31, t(314) = −4.62, p = .001; Table 3). These results show that, independent of psychological health, perceiving parental praise to be inaccurate predicted low school performance for children.

Table 3.

Regression Analysis Predicting the Grade Point Average after Controlling for Depression.

Variable	B	SE	t	p	Partial Correlation [95% CI]
Depression	−0.31	0.07	−4.62	<.01	−.25 [−.35, −.15]
PAC	1.38	1.07	1.30	.20	.07 [−.04, .18]
PAC²	−0.98	0.55	−1.80	.07	−.10 [−.21, .01]
R ²	.08

Note. PAC = praise accuracy perceived by child; CI = confidence interval.

Further Testing the Robustness of the Findings

Next, we controlled for some potential covariates to assess the robustness of our findings (see Table 4). Looking at the amount of parental praise for their children (5.11 ± 1.08), we found that it did not predict the children’s GPA (B = 0.42, t(284) = 0.84, p = .40) or depression score (B = −0.49, t(284) = −1.19, p = .24). After controlling for this variable, the quadratic effect of a parent’s perceived praise accuracy was still reliable for GPA (B = −0.73, t(282) = −2.28, p = .02) and depression (B = 0.79, t(282) = 2.93, p = .004). How much parents monitored their children’s academic performance (4.70 ± 1.11) predicted the GPA (B = 1.21, t(284) = 2.58, p = .01), with higher monitoring being associated with higher GPA levels, but not with depression (B = −0.16, t(284) = −0.40, p = .69). After controlling for this variable, the observed quadratic effect was still reliable for both GPA (B = −0.75, t(282) = −2.35, p = .02,) and depression (B = 0.76, t(282) = 2.83, p = .01).

Table 4.

Regression Analyses Predicting the Grade Point Average or Depression Score After Controlling for Third Variables.

Variable	Models Predicting GPA Controlling for
	Amount of Praise		Parent’s Monitoring on Academic Performance		Mother’s Education Level		Father’s Education Level		Household Income
	B (SE)	Partial Correlation [95% CI]	B (SE)	Partial Correlation [95% CI]	B (SE)	Partial Correlation [95% CI]	B (SE)	Partial Correlation [95% CI]	B (SE))	Partial Correlation [95% CI]
Control variable	1.38 (.56)**	.15 [.03, .26]	1.36 (.47)***	.17 [.06, .28]	1.99 (.87)**	.14 [.02, .25]	0.54 (.78)	.04 [−.08, .16]	0.82 (.73)	.07 [−.05, .18]
PAP	−0.92 (.60)	−.09 [−.21, .03]	−0.35 (.55)	−.04 [−.15, .08]	−0.44 (.56)	−.05 [−.16, .07]	−0.40 (.56)	−.04 [−.16, .07]	−0.45 (.57)	−.05 [−.16, .07]
PAP²	−0.73 (.32)**	−.13 [−.25, −.02]	−0.75 (.32)**	−.14 [−.25, −.02]	−0.62 (.33)**	−.11 [−.23, .00]	−0.65 (.33)**	−.12 [−.23, .00]	−0.61 (.33)*	−.11 [−.23, .00]
R ²	.054		.062		.049		.033		.035
	Models Predicting Depression Controlling for
	Amount of Praise		Parent’s Monitoring on Academic Performance		Mother’s Education Level		Father’s Education Level		Household Income
Variable	B (SE)	Partial Correlation [95% CI]	B (SE)	Partial Correlation [95% CI]	B (SE)	Partial Correlation [95% CI]	B (SE)	Partial Correlation [95% CI]	B (SE)	Partial Correlation [95% CI]
Control variable	−0.99 (.47)**	−.12 [−.24, −.01]	−0.26 (.40)	−.04 [−.15, .08]	0.89 (.72)	.07 [−.04, .19]	0.69 (.64)	.06 [−.05, .18]	−0.70 (.61)	−.07 [−.18, .05]
PAP	−0.01 (.50)	−.001 [−.12, .12]	−0.43 (.46)	−.05 [−.17, .06]	−0.51 (.47)	−.06 [−.18, .05]	−0.49 (.46)	−.06 [−.18, .05]	−0.41 (.47)	−.05 [−.17, .06]
PAP²	0.79 (.27)***	.17 [.06, .28]	0.76 (.27)***	.17 [.05, .28]	0.79 (.27)***	.17 [.06, .28]	0.77 (.27)***	.17 [.05, .28]	0.76 (.27)***	.17 [.05, .28]
R ²	.045		.031		.036		.035		.035

Note. Control variables include amount of praise, parental monitoring of academic performance, maternal education level, paternal education level, and household income. PAP = praise accuracy perceived by parent.

*p < .10. **p ≤ .05. ***p < .01.

The maternal education level predicted the GPA (B = 1.97, t(284) = 2.26, p = .03) of their children, but not the depression scores (B = 0.83, t(283) = 1.14, p =.26). When controlling for this, the quadratic effect was still reliable for both GPA (B = −0.62, t(282) = −1.90, p = .05) and depression (B = 0.79, t(281) = 2.92, p = .004). By contrast, neither the paternal education level nor the household income predicted GPA or depression in children (ts < 2.00, ns). However, when controlling for the paternal education level, the quadratic effect was still reliable for both GPA (B = −0.65, t(283) = −1.99, p = .04) and depression (B = 0.77, t(282) = 2.86, p = .01). Finally, controlling for household income, the quadratic effect was again reliable for both GPA (B = −0.61, t(281) = −1.88, p = .06) and depression (B = 0.76, t(280) = 2.81, p = .01).

Testing Alternative Accounts

Our findings provide strong evidence that receiving praise perceived to be accurate concerning their schoolwork was associated with positive outcomes for children. However, alternative interpretations are possible given the correlational nature of this study; this includes the possibility that causality may actually run in the opposite direction. In other words, praise perceived to be accurate, by both parents and children, could be more a consequence of high achievement and less a cause of it. Parents, for example, could find it easier to offer accurate praise to high-performing children, but struggle to offer accurate or appropriate praise to low-performing children. To evaluate the validity of this alternative perspective, we tested whether the observed quadratic effect of parents’ perceived praise accuracy on the children’s GPA remained reliable after controlling for their child’s academic aptitude, as captured by their performance (GPA) in the previous semester. As expected, the GPA in the previous semester predicted their GPA in the current semester (B = 0.99, t(320) = 46.87, p = .001). Moreover, the GPA in the previous semester also predicted the parental perceptions of praise accuracy, as measured by the absolute value of the difference between the parental perceptions of praise accuracy and the midpoint of the scale (B = −0.01, t(285) = −2.40, p = .02), and indicating that high-performing children were more likely to receive praise perceived to be accurate from their parents compared with low-performing children. However, even after controlling for the effect of the previous semester’s GPA, the quadratic effect of parental perceived praise accuracy was still significant for the GPA in the current semester (B = −2.78, t(279) = −2.90, p = .004).

It is also possible that the observed quadratic effect of the children’ perceptions of praise accuracy on their GPA was driven by their own academic aptitude, such that high-performing children potentially did not receive or did not perceive that they received praise that exceeded reality. We tested whether the observed quadratic effect of children’s perceived praise accuracy would remain reliable on their GPA after controlling for their academic aptitude, as captured by their GPA in the previous semester. The children’s past GPA marginally predicted their perceptions of praise accuracy, as measured by the absolute value of the difference between their perceptions of praise accuracy and the scale midpoint (B = −0.01, t(314) = −1.76, p = .08). Consistent with the pattern seen with parental perceptions, high-performing children were more likely to perceive that they received accurate praise from their parents compared with low-performing children. However, even after controlling for the effect of the previous GPA, the quadratic effect of children’s perceived praise accuracy on their current GPA was still significant (B = −7.17, t(309) = −2.58, p = .01; Table 5).

Table 5.

Regression Analyses Predicting the Grade Point Average (GPA) After Controlling for the Previous Semester’s GPA.

Variable	B	SE	t	p	Partial Correlation [95% CI]	Variable	B	SE	t	p	Partial Correlation [95% CI]
Previous GPA	0.93	0.36	25.85	<.01	.84 [.80, .87]	Previous GPA	0.98	0.27	36.38	<.01	.90 [.88, .92]
PAP	0.91	1.61	0.57	.57	.03 [−.08, .15]	PAC	8.86	3.87	2.29	.01	.13 [.02, .24]
PAP²	−2.78	0.96	−2.90	.00	−.17 [−.28, −.06]	PAC²	−7.17	2.78	−2.58	.01	−.14 [−.25, −.03]
PAP × previous GPA	−0.01	0.02	−0.66	.51	−.04 [−.16, .08]	PAC×previous GPA	−0.08	0.04	−1.91	.06	−.11 [−.22, .00]
PAP² ×previous GPA	0.03	0.01	2.82	.01	.17 [.05, .28]	PAC²×previous GPA	0.07	0.03	2.34	.02	.13 [.02, .24]
R ²	0.87					R ²	0.87

Note. PAP = praise accuracy perceived by parent; PAC = praise accuracy perceived by child.

These results suggest that even though children with high academic aptitude elicit praise that is perceived to be more accurate (both by the parents and by the children themselves), the positive outcomes associated with praise perceived to be accurate hold, even when controlling for the academic aptitude of children.

Discussion and Concluding Remarks

Praise is still one of the most common tools used when parenting and educating children. Many parents hold beliefs based on folk theories that praise is beneficial and use praise to motivate children in learning and to help them build self-esteem and self-confidence. However, research evidence for the presumed benefits of praise is mixed. In the present study, we attempted to explore this issue by showing how praise could be both beneficial and detrimental to children depending on the perceived accuracy of the praise, as judged both by givers (the parents) and by recipients (the children).

We demonstrated that when parents perceived that they over- or underpraised their children for schoolwork, children performed worse in school and experienced depression to a greater extent, as compared with children whose parents thought their praise accurately reflected reality. How much the children perceived their parents to over- or underpraise them had analogous effects, with the exception that children’s perceptions of slightly overstated praise were as beneficial as accurate praise. We also found that, independent of parental perceptions, when children perceived their parents to give over- or underpraise, they suffered from poor academic performance and emotional distress. Together, these findings underscore the benefits of giving praise to children that closely reflects their actual performance, and of delivering it in a way they perceive to be accurate.

The present research particularly highlights that praise, just like feedback, should be understood as an interactive process, with consideration given to how it is perceived, accepted, and responded to by the recipient (Ilgen, Fisher, & Taylor, 1979). In particular, our study shows that, independent of how accurate the praise is, the way it is delivered and perceived has significant implications and consequences for academic achievement and psychological well-being among children. Emphasizing this point, we further identified that the accuracy of praise, as perceived by the children, affected their academic performance and well-being beyond their parents’ accuracy perceptions.

An important contribution of this study is that the findings will benefit parents and educators by offering valuable guidance on how to praise children and students effectively. With the self-esteem movement that began in the United States in the 1960s, spreading the conviction that viewing oneself positively was beneficial, even when not necessarily based on reality, many teachers and parents believe that praise is beneficial to academic achievement and psychological well-being (Miller, Wang, Sandel, & Cho, 2002; Seligman, Reivich, Jaycox, & Gillham, 1995). However, our study suggests that parents and educators should pay attention to how they feel about the accuracy of the praise they give children and students. As our results reveal, self-admitted gaps between given praise and actual performance had predictive power for important child outcomes. If parents and educators feel uneasy when they think about the accuracy of the praise they offer, they should take this as a warning sign. Furthermore, our findings emphasize the importance of children’s perceptions over those of their parents. Instigating “heart-to-heart” talks with children about whether they feel adequately praised could be one way to address existing problems and improve the psychological and academic outcomes.

In the current study we tried, to the best of our ability, to control for variables that could potentially have affected the results, and to test the justifiability of alternative accounts. Even though our results withstood these tests of robustness, the fact remains that this was a correlational study that precludes claims of causality. Our ability to eliminate alternative explanations, particularly concerning the direction of effects and third variables, was limited as a consequence. Although we consider it unlikely, it therefore remains possible that the causal direction may be the reverse of that proposed. It is also possible that some third variable, such as the quality of the parent–child relationship, led both to perceptions of accurate versus inaccurate praise and to positive versus negative child outcomes.

Despite the limitations of this work, we hope that the current study will inform and inspire future studies to test more complex models of the effects of praise accuracy on outcomes in children. It would also be fruitful for future studies to delve deeper into the question of what it means for parents to give accurate, overstated, or understated praise, because this may illuminate the specific practices associated with each approach. For example, might accurate praise differ from inaccurate praise in its frequency, extremity, content, or quality? A substantive examination of praise accuracy could provide information on the actual practices that distinguish accurate praise from inaccurate praise.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Research Foundation of Korea (NRF) Grant funded by the Korea government (MEST; No. 2012S1A5A8023903).

Notes

References

Barker

G. P.

Graham

(1987). Developmental study of praise and blame as attributional cues. Journal of Educational Psychology, 79, 62–66.

Baumeister

R. F.

Hutton

D. G.

Cairns

K. J.

(1990). Negative effects of praise on skilled performance. Basic and Applied Social Psychology, 11, 131–148. doi:10.1207/s15324834basp1102_2

Beck

A. T.

Steer

R. A.

Brown

G. K.

(1996). Beck Depression Inventory-II. San Antonio, TX: The Psychological Corporation.

Brummelman

Thomaes

de Castro

B. O.

Overbeek

Bushman

B. J.

(2014). “That’s not just beautiful—that’s incredibly beautiful!” The adverse impact of inflated praise on children with low self-esteem. Psychological Science, 728–735. doi:10.1177/0956797613514251

Cameron

Pierce

W. D.

(1994). Reinforcement, reward, and intrinsic motivation: A meta-analysis. Review of Educational Research, 64, 363–423. doi:10.3102/00346543064003363

Cimpian

Arce

H. M. C.

Markman

E. M.

Dweck

C. S.

(2007). Subtle linguistic cues affect children’s motivation. Psychological Science, 18, 314–316. doi:10.1111/j.1467-9280.2007.01896.x

Delin

C. R.

Baumeister

R. F.

(1994). Praise: More than just social reinforcement. Journal for the Theory of Social Behaviour, 24, 219–241. doi:10.1111/j.1468-5914.1994.tb00254.x

Garner

P. W.

(2006). Prediction of prosocial and emotional competence from maternal behavior in African American preschoolers. Cultural Diversity and Ethnic Minority Psychology, 12, 179–198. doi:10.1037/1099-9809.12.2.179

Gunderson

E. A.

Gripshover

S. J.

Romero

Dweck

C. S.

Goldin-Meadow

Levine

S. C.

(2013). Parent praise to 1- to 3-year-olds predicts children’s motivational frameworks 5 years later. Child Development, 84, 1526–1541. doi:10.1111/cdev.12064

10.

Grusec

J. E.

(1991). Socializing concern for others in the home. Developmental Psychology, 27, 338–342. doi:10.1037/0012-1649.27.2.338

11.

Harari

Covington

M. V.

(1981). Reactions to achievement behavior from a teacher and student perspective: A developmental analysis. American Educational Research Journal, 18, 15–28.

12.

Hastings

P. D.

McShane

K. E.

Parker

Ladha

(2007). Ready to make nice: Parental socialization of young sons’ and daughters’ prosocial behaviors with peers. The Journal of Genetic Psychology, 168, 177–200. doi:10.3200/GNTP.168.2.177-200

13.

Henderlong

Lepper

M. R.

(2002). The effects of praise on children’s intrinsic motivation: A review and synthesis. Psychological Bulletin, 128, 774–795. doi:10.1037/0033-2909.128.5.774

14.

Ilgen

D. R.

Fisher

C. D.

Taylor

M. S.

(1979). Consequences of individual feedback on behavior in organizations. Journal of Applied Psychology, 64, 349–371.

15.

Judd

C. M.

McClelland

G. H.

Ryan

C. S.

(2011). Data analysis: A model comparison approach. New York, NY: Routledge.

16.

Kamins

M. L.

Dweck

C. S.

(1999). Person versus process praise and criticism: Implications for contingent self-worth and coping. Developmental Psychology, 35, 835–847. doi:10.1037/0012-1649.35.3.835

17.

Kast

Connor

(1988). Sex and age differences in responses to informational and controlling feedback. Personality and Social Psychology Bulletin, 14, 514–523. doi:10.1177/0146167288143010

18.

Kim

Y. H.

Chiu

C. Y.

(2011). Emotional costs of inaccurate self-assessments: Both self-effacement and self-enhancement can lead to dejection. Emotion, 11, 1096–1104. doi:10.1037/a0025478

19.

Kim

Y. H.

Chiu

C. Y.

Zou

(2010). Know thyself: Misperceptions of actual performance undermine achievement motivation, future performance, and subjective well-being. Journal of Personality and Social Psychology, 99, 395–409. doi:10.1037/a0020555

20.

Kinicki

A. J.

Prussia

G. E.

B. J.

McKee-Ryan

F. M.

(2004). A covariance structure analysis of employees’ response to performance feedback. Journal of Applied Psychology, 89, 1057–1069. doi:10.1037/0021-9010.89.6.1057

21.

Kohn

(1993). Punished by rewards: The trouble with gold stars, incentive plans, A’s, praise, and other bribes. New York, NY: Houghton Mifflin.

22.

Kun

Parsons

J. E.

(1974). Development of integration processes using ability and effort information to predict outcome. Developmental Psychology, 10, 721–732.

23.

Lepper

M. R.

Greene

D. E.

(1978). The hidden costs of reward: New perspectives on the psychology of human motivation. Oxford, England: Lawrence Erlbaum.

24.

Meyer

W.-U.

Bachmann

Biermann

Hempelmann

Ploger

F.-O.

Spiller

(1979). The informational value of evaluative behavior: Influences of praise and blame on perceptions of ability. Journal of Educational Psychology, 71, 259–268.

25.

Miller

A. T.

Hom

H. L.

(1997). Conceptions of ability and the interpretation of praise, blame, and material rewards. Journal of Experimental Education, 65, 163–177.

26.

Miller

P. J.

Wang

S. H.

Sandel

Cho

G. E.

(2002). Self-esteem as folk theory: A comparison of European American and Taiwanese mothers’ beliefs. Parenting: Science and Practice, 2, 209–239.

27.

Mueller

C. M.

Dweck

C. S.

(1996, 04). Implicit theories of intelligence: Relation of parental beliefs to children’s expectations. Poster session presented at Head Start’s Third National Research Conference, Washington, DC.

28.

Mueller

C. M.

Dweck

C. S.

(1998). Praise for intelligence can undermine children’s motivation and performance. Journal of Personality and Social Psychology, 75, 33–52. doi:10.1037/0022-3514.75.1.33

29.

Murayama

Pekrun

Suzuki

Marsh

H. W.

Lichtenfeld

(2016). Don’t aim too high for your kids: Parental overaspiration undermines students’ learning in mathematics. Journal of Personality and Social Psychology, 111, 766–779.

30.

Nicholls

J. G.

(1978). The development of the concepts of effort and ability, perception of academic attainment, and the understanding that difficult tasks require more ability. Child Development, 49, 800–814.

31.

Pomerantz

E. M.

Kempner

S. G.

(2013). Mothers’ daily person and process praise: Implications for children’s theory of intelligence and motivation, Developmental Psychology, 49, 2040–2046. doi:10.1037/a0031840

32.

Reid

B. V.

Valsiner

(1986). Consistency, praise, and love: Folk theories of American parents. Ethos, 14, 282–304.

33.

Seligman

M. E.

Reivich

Jaycox

Gillham

(1995). The optimistic child. Boston, MA: Houghton, Mifflin, & Co.

34.

Taylor

M. S.

Fisher

C. D.

Ilgen

D. R.

(1984). Individual’s reactions to performance feedback in organizations: A control theory perspective. In Rowland

K. M.

Ferris

G. R.

(Eds.), Research in personnel and human resources management (Vol. 2, pp. 81–124). Greenwich, CT: JAI Press.

35.

Weiner

(1992). Human motivation: Metaphors, theories, and research. Newbury Park, CA: Sage.