Stability during Childhood

Longitudinal studies, examinations of the same groups of individuals over time, are costly and dif ficult to conduct. As a result, there are precious few of such studies to draw on. A major exception is the Block and Block Longitudinal Study , which initiated the testing of a sample of more than 100 children from the Berkeley-Oakland area of California when the children were merely 3 years old (see, e.g., Block & Robbins, 1993). Since that time, the sample has been followed and repeatedly tested at ages 4, 5, 7, 1 1, and into adulthood.

One of the first publications from this project focused on individual di ferences in activity level (Buss, Block, & Block, 1980). When the children were 3 years old, and then again at 4, their activity levels were assessed in two ways. The first wa through the use of an actometer, a recording device attached to the wrists of the children during several play periods. Motoric movement activated the recording device— essentially a self-winding wristwatch. Independently, the children's teachers completed ratings of their behavior and personalities. The behavioral measure of activity level contained three items that were directly relevant: "is physically active," "is vital, energetic, active," and "has a rapid personal tempo." These items were summed to form a total measure of teacher -observed activity level. This observer -based measure was obtained when the children were 3 and 4 and then again when they reached age 7.

Table 5.2 shows the correlations among the activity level measures, both at the same ages and across time to assess the stability of activity level during childhood. The correlations between the same measures obtained at two dif ferent points in time are called stability coefficient (these are also sometimes called test-retest reliability coefficients). The correlations between dif ferent measures of the same trait obtained at the same time are called validity coefficients

Several key conclusions about validity and stability can be drawn from T able 5.2. First, notice in T able 5.2 that the actometer -based measurements of activity level have significant positive validity coe ficients with the judge-based measurements of activit

Table 5.2 Intercorrelations among Activity Measures

A C T O M

E T E R

J U D G

E - B A S

E D

Age 3

Age 4

Age 3

Age 4

Age 7

Actometer:

Age 3

.44*

.61***

.56***

.19

Age 4

.43**

.66***

.53***

.38**

Judge-based:

Age 3

.50***

.36**

.75***

.48***

Age 4

.34*

.48***

.51***

.38**

Age 7

.35*

.28*

.33*

.50***

*p < .05. **p < .01. ***p < .001 (two-tailed). Correlations above the ellipses (. . .) are based on boys' data, those below the ellipses (. . .) are based on girls' data.

*p < .05. **p < .01. ***p < .001 (two-tailed). Correlations above the ellipses (. . .) are based on boys' data, those below the ellipses (. . .) are based on girls' data.

Figure 5.1

The figure shows the stability of aggression in males over di ferent time intervals. Aggression shows the highest levels of stability over short time intervals such as from one year to the next. As the time interval between testings increases, however, the correlation coefficients decline, suggesting that aggressivenes changes more over long time intervals than over short time intervals.

Figure 5.1

The figure shows the stability of aggression in males over di ferent time intervals. Aggression shows the highest levels of stability over short time intervals such as from one year to the next. As the time interval between testings increases, however, the correlation coefficients decline, suggesting that aggressivenes changes more over long time intervals than over short time intervals.

level. Activity level in childhood can be validly assessed through both observational judgments and activity recordings from the actometers. The two measures are moderately correlated at each age, providing cross-validation of each type of measure.

Second, notice that the correlations of the activity level measurements in T able 5.2 are all positively correlated with measurements of activity level taken at later ages. We can conclude from these correlations that activity level shows moderate stability during childhood. Children who are highly active at age 3 are also likely to be active at ages 4 and 7. Their less active peers at age 3 are likely to remain less active at ages 4 and 7.

Finally, notice that the size of the correlations in T able 5.2 tend to decrease as the time interval between the dif ferent testings increases. This finding parallels th finding about infancy made by Rothbart (1986). As a general rule, the longer the time between testings, the lower the stability coef ficients. In other words, measures take early in life can predict personality later in life, but the predictability decreases with the length of time between the original testing and the behavior being predicted.

These general conclusions apply to other personality characteristics as well. Aggression and violence have long been a key concern of our society . In recent years in the United States, violence has captured the attention of the whole country . For example, the startling killings by two students at Columbine High School shocked the country. These and other similar shootings have prompted many to ask, "What causes some children to act so aggressively?"

As it turns out, numerous studies of childhood aggression have been conducted by personality psychologists. Dan Olweus (1979) reviewed 16 longitudinal studies of aggression during childhood. The studies varied widely on many aspects, such as age at which the children were first tested (2-18), length interval between first testing a final testing (half a year to 18 years), and the specific measures of aggression us (e.g., teacher ratings, direct observation, and peer ratings).

Figure 5.1 shows a summary graph of the results of all these studies. The graph depicts the stability coefficients for aggression as a function of the interval between fir

Eliminating Stress and Anxiety From Your Life

Eliminating Stress and Anxiety From Your Life

It seems like you hear it all the time from nearly every one you know I'm SO stressed out!? Pressures abound in this world today. Those pressures cause stress and anxiety, and often we are ill-equipped to deal with those stressors that trigger anxiety and other feelings that can make us sick. Literally, sick.

Get My Free Ebook


Post a comment