Intraindividual variation among pregnant adolescents: a pilot study and conceptual discussion.

Blinn-Pike, Lynn M. ; Stenberg, Lori ; Thompson, Cecelia 等

Background

After many years of working with, researching, and reading survey research about pregnant adolescents, the conclusion was reached that samples of pregnant adolescents provide self-report data of questionable psychometric value. First, they display very labile moods. Second, they may experience cognitive-developmental changes due to the experience of being pregnant (Blinn, 1988). Third, they tend to provide inaccurate information about areas such as family or parental socioeconomic status, and tend toward social desirability when it comes to self-report data on sensitive information such as substance use, sexual behavior, and delinquent behavior. An adolescent who is one month pregnant is quite different in her thoughts, attitudes, and expectations from one who is further along in her pregnancy, even by a few weeks. A pregnant adolescent's responses at one time may be very different one hour, day, week or month later. To report which trimester of pregnancy an adolescent is in does not recognize changes that occur on a more frequent basis. The purpose of the present pilot study was to illustrate the lability of pregnant adolescents' moods which may impact on both their decision-making processes and the psychometric properties of the data they provide.

Researchers in the area of adolescent pregnancy have several unique problems, and researchers who want to do short-term longitudinal work and address changes over the course of the pregnancy have an even more difficult task. Gaining access to samples of pregnant adolescents can be difficult. Samples of pregnant adolescents are generally heterogeneous in terms of gestational progress, life situation, and age of the adolescents. Recognizing these sampling issues as unavoidable in this area of research, it is imperative that researchers compensate by reporting more detail about the gestational variation in the samples, at least the months of the subjects' pregnancies. They also need to more critically address the issues of the validity and reliability of their data.

To confirm the conclusion that there was a lack of specificity in the gestational and psychometric data provided on survey research with pregnant adolescents, an informal content analysis was conducted using 30 research articles published in refereed journals from 1977 to the present dealing with the measurement of psychosocial variables in pregnant adolescents. All of the studies were selected because they used survey methodologies. Two methodological patterns were revealed. First, for understandable and practical reasons, all of the research involved convenience samples with adolescents in various stages of pregnancy. However, little specific information was provided about gestational progress other than by trimesters. Common phrases used to describe the pregnancy status of the adolescent samples were: "All of them were past the first trimester." "All were in the second trimester or beyond." "All were in the end of their first trimester or in the second trimester when they entered the study." "All were at least four weeks from delivery." None of the authors reported the mean and standard deviation of the pregnancy status of the sample or recognized that there may be variations due to finer distinctions in pregnancy status.

Second, none of the researchers reported the test-retest or split-half reliabilities of the instruments used based on the pregnant samples being studied. None addressed validity issues by including measures of social desirability. Those authors who included information on the reliability of the survey instruments they used reported reliability coefficients previously determined with nonpregnant samples. The instruments used were all well known for their application with nonpregnant samples and included the Piers-Harris Self-Concept Questionnaire (Piers & Harris, 1983). Rosenberg Self-Esteem Questionnaire (Rosenberg, 1965), Offer Self-Image Questionnaire for Adolescents (Offer, Ostrov, & Howard, 1982), State-Trait Anxiety Inventory (Spielberger, Gorsuch, & Lushene, 1970), Beck Depression Inventory (Beck et al., 1979), and the Tennessee Self-Concept Scale (Fitts, 1965).

The findings led to the present attempt to illustrate variations in pregnant adolescents' moods. The intent was not to provide generalizable findings about the moods of pregnant adolescents, but to provide rich, qualitative data on a small sample of pregnant adolescents in order to document the need to explore this area in greater depth in future research.

Nesselroade (1991, p. 6) makes a distinction in developmental research between studying intraindividual variation and intraindividual change. Intraindividual variation is defined as "short-term and often reversible changes in emotions," such as moods or states in the state-trait distinction. Intraindividual change is defined as long-term and irreversible changes often associated with development, learning, and progressive organic changes. This is related to the trait element of the state-trait distinction. Intraindividual variability among pregnant adolescents was the specific focus of the present effort. Intraindividual variability may be overlooked as a threat to the reliability of much self-report survey data and has been virtually ignored in research with pregnant adolescents. The reliability of survey research with pregnant adolescents needs to be reexamined in light of the increasing evidence that self-report data from this population are influenced by intraindividual variations.

Specifically, the foci of this study were the affective tone, contextuality, and lability of pregnant adolescents' moods over time. Mood was used interchangeably with intraindividual variation and was defined as a quickly changing, intensely felt, and wildly fluctuating emotional experience (Bence, 1992). Lability and affective tone have been the main variables addressed in the few research studies that have been done on adolescent moods (Bence, 1990, 1992). The present work was unique in that it added the dimension of contextuality, which was defined as relating to three topics: their feelings about their mothers, babies' fathers, and selves.

Traditionally, mood states have been viewed as easily influenced by situational factors, whereas feelings about the self and others have been considered more stable and enduring. This study questioned whether pregnant adolescents' mood states may, at least temporarily, influence how they feel about themselves and those around them. It also questioned whether the influence of emotional highs or lows are generalizable across their feelings toward the people around them. For example, if they are in a very negative mood, does this feeling carry over to their feelings toward themselves, their mothers, and their babies' fathers?

LITERATURE REVIEW

It has come to be accepted as part of the popular culture that both pregnant women and nonpregnant adolescents experience mood swings, including extreme positive and negative emotional states. The labile feelings of these two groups have been attributed to both psychological and hormonal causes. Yet there has been no research on the variations in moods among pregnant adolescents, the focus of this investigation. In one of the few studies examining intraindividual change among pregnant adolescents, Blinn (1988) examined how pregnant adolescents' views of their future changed over the course of their pregnancies. She reported that their expectations changed from the second to the third trimester of pregnancy. Blinn asked that closer attention be paid to a possible cognitive-developmental theory of adolescent decision making during pregnancy.

Due to the fact that there is no research on the mood states of pregnant adolescents, the literature was reviewed in two separate areas: mood variations during adolescence, and during pregnancy. No research has been done, and it is beyond the scope of this paper, to determine if mood swings during adolescence compound "normal" mood swings during pregnancy.

Intraindividual Variations During Adolescence

Although the "storm and stress" hypothesis of adolescence is no longer strongly supported, adolescence may still be a time of heightened emotional state (Csikszentmihalyi & Larson, 1987; Larson, Csikszentmihalyi, & Graef, 1980). Adolescents appear to experience more intense feelings than do adults (Bradburn 1969, Campbell 1981; Diener, Sandvik, & Larsen, 1985). Larson, Csikszentmihalyi, and Graef (1980) used electronic pagers to alert adolescents and adults to record their emotional feelings at random times during the day. They found that adolescents reported more occasions of both negative and positive feelings.

According to Brooks-Gunn and Reiter (1990, p. 43), "Perhaps the most commonly held belief about puberty is that it heralds increased emotionality in the form of negative moods and swings in moods. While not as intense or extreme as popular accounts would have us believe, on average, increases in negative emotions do seem to occur between late childhood and late adolescence. Whether mood swings increase as well is a fascinating and virtually unexplored issue."

Research on adolescent moodiness should not be ignored because the storm and stress view of adolescence has been under fire. Buchanan (1991) stressed the importance of studying adolescent moods by hypothesizing that the challenge to the storm and stress stereotype of adolescence has been based on research on such behaviors as conflict with parents and deviance, rather than on research examining moods. It is possible that moods become more unstable at puberty, but that they do not bring on crises in the adolescents who are experiencing them.

Intraindividual Variations During Pregnancy

It is important to understand mood variability during pregnancy because there appear to be strong relationships between pregnancy mood states and (a) intrapartum processes such as labor and delivery complications (Perez, 1983), (b) intrapartum analgesia requirements (Perez, 1983), and (c) physical and psychological postpartum adjustment (Smith, Cubis, Brinsmead, & Levin, 1990). Likewise, the effects of depressed maternal mood have been shown to lessen the ability of mothers to elicit positive responses from their infants, and of infants to contingently respond to their mothers (Zekoski, O'Hara, & Willis, 1987).

Smith et al. (1990) investigated the relationships among mood changes, obstetric experience, and hormonal changes in 97 primiparous women. Smith found that mood disturbances rose significantly during pregnancy and peaked at birth, as did levels of plasma cortisol, beta-endorphin, and corticotrophin-releasing hormones. He also found that the women who had the highest mood disturbance scores at 28 weeks gestation received significantly more pain relief during delivery.

Ballinger, Kay, Naylor, and Smith (1982) found that women who had fewer negative moods during pregnancy also reported more positive mood changes two to four days after delivery. They further stated that women showing a positive mood change after delivery were distinct in both mood and biochemistry during pregnancy, and this may help identify these women prior to delivery. Buchanan (1991) traced women's emotional states at three and eight months of pregnancy, and again at three and eight months postpartum. High emotional lability was found to be positively associated with postpartum depression.

Finally, Salovey and Birnbaum (1989) reported that the mood status of ill persons in general may negatively influence their perceptions of levels of control over their health status and ability to carry out health-seeking behaviors. This may result in failure to follow prescribed regimens and to retard recovery. Although the Salovey and Birnbaum research did not involve pregnant women, it may be that the negative moods experienced by some pregnant women influence health and lifestyle behaviors which in turn affect the health and well-being of their babies.

Given the critical nature of the decisions made by pregnant adolescents, and the implications drawn from their self-report data, it is important to not only chart their mood changes, but to explore the context of these moods. In this study, in order to explore the context-specific nature of mood states, fluctuations in pregnant adolescents' feelings about their mothers, babies' fathers, and selves were compared. The pregnant adolescents' feelings about their mothers and babies' fathers were considered important because they may be the people who are closely involved in the situation and who have a substantial impact on decisions concerning pregnancy outcome, marriage, education, and health-related behaviors.

In addition, pregnant adolescents' mothers and babies' fathers were selected because of the conflicting findings concerning their relationships with these two individuals. Early researchers reported that an adolescent pregnancy worsens the mother-daughter relationship (Crumidy & Jacobziner, 1966; Friedman, 1966). Others have reported that it strengthens the relationship (Smith, 1975; Young, Berkman, & Rehr, 1975). A number of research studies have shown that the involvement of the babies' fathers during pregnancy improves adolescent mothers' sense of competency, self-image, and attachment to their babies (Cervera, 1991; Panzarine, 1983). On the other hand, Giblin, Poland, and Sams (1987) reported that tangible and emotional support from adolescent fathers was not related to the pleasure the pregnancy provided for the adolescents. Pleasure with pregnancy was positively associated with assistance from their mothers and satisfaction with their living arrangements.

Little is known about the interaction between pregnant girls' feelings toward their mothers and their babies' fathers. It is not known if a negative relationship with one of them makes the pregnant adolescent feel particularly vulnerable and overly dependent on the other. Likewise it is not known if the affective tone of one relationship generalizes to the other relationship.

Guiding Questions

The purpose here was to begin to illustrate and understand intraindividual variations in pregnant adolescents. The following questions guided this effort: What mood patterns (affective tone and lability) are evident within the context of pregnant adolescents' feelings about themselves, their mothers, and their babies' fathers over time? How are their mood patterns concerning themselves, their mothers, and their babies' fathers related? What tentative hypotheses can be generated about interindividual variability during adolescent pregnancy from examining multiple cases of intraindividual variability?

METHOD

Participants

The participants consisted of 14 pregnant adolescents who volunteered to keep weekly written diaries concerning their feelings about themselves, their mothers, and their babies' fathers for six weeks. They were all attending regular classes in their public high schools and were asked to volunteer for the project by their classroom teachers. The diaries were completed on a weekly schedule because there is evidence that daily fluctuations in mood exhibit a seven-day cycle based on social, psychological, and biological processes (see Larsen & Kasimatis, 1990). All of the adolescents lived in Idaho and were each paid ten dollars for their involvement.

All of the females were white with a mean age of 16 years (SD = 1.18). The girls' levels in school were: 9th grade (2), 10th grade (3), 11th grade (5), and 12th grade (4). Nine of them were receiving at least one form of public financial assistance such as food stamps or AFDC. The mean number of months pregnant was 5.1 (SD = 1.28). Their pregnancy status was: three months pregnant (2), four months (2), five months (3), six months (6), and seven months (1).

Two were married, the others had never been married. Eight lived with at least one of their parents, two lived with their husbands, one lived with a sister, and three lived in foster or residential homes. Most of them were lower or lower middle class as evidenced by the places of employment and the educational levels of their parents. One father had completed elementary school, six had completed high school, two had undergraduate degrees, and two had other forms of vocational or postsecondary training. Three females provided no information about their fathers' level of education. Three mothers had completed elementary school, seven had completed high school or the GED, one had an undergraduate degree, and two had other forms of postsecondary education. One girl did not provide information about her mother's educational level.

Instrumentation

Bence (1990) stated that little is known about the developmental aspects of mood variability among nonclinical adolescents because of problems with the measurement of mood variability. Mood variation is most often measured by calculating the standard deviation of mood self-reports. The adolescents are often asked to rate their emotional moods on a form of semantic differential or to select from a series of adjective pairs. Measurement includes calculating the mean and standard deviation of each adolescent's series of mood reports (see Larson & Lampman-Petraitis, 1989 for a more detailed discussion).

Another technique for measuring rate of mood change includes spectral analysis or Fast Fournier transformation of each adolescent's time-ordered composite mood scores. This procedure uses sine and cosine waves of different frequencies to describe any cyclical phenomena that may be occurring (see Larsen, 1987; Larsen & Kasimatis, 1990).

Although these procedures clearly trace the patterns of adolescents' moods, due to the present interests in the qualitative nature and contextuality of moods, a creative methodology was employed. All 14 females were asked to write about their feelings toward their mothers, babies' fathers, and selves on a weekly basis for six weeks. Their teachers collected the diaries each week and mailed them to the researchers. The adolescents were each given a form with the same type of fill-in-the-blank question for each topic (mother, baby's father, or self): "During the last week I thought |my mother, my baby's father, or I~ was . . . because. . . ." In addition, during the first week of participation, their teachers assisted them in completing demographic information sheets.

Data Analysis

Four individuals read each diary and provided 18 ratings per diary. They were asked to read each entry and rate their overall impression without further training. They rated each of the three topics for each of the six weeks on a five-part scale with one (very negative) to five (very positive) reflecting the affective tone of each diary entry. Three of the raters were faculty members in departments of Home Economics or Child and Family Studies at universities in Idaho, Tennessee, and Arkansas. One rater was a graduate student. Each rater worked independently, and there was no training or discussion of the content of the diaries prior to rating. Data analysis involved the calculation of interrater reliabilities, descriptive statistics, and correlations.

RESULTS

The reliabilities for each topic (mother, baby's father, self) were calculated for the four raters for each of the six weeks. The Spearman Brown reliability coefficient for mother was .95; for baby's father, .94; and for self, .90. The reliabilities were considered high enough to justify reporting the average of the four ratings as indicators of affective tone for each of the three topics in the remainder of the results.

The means, standard deviations, and ranges of the raters' scores were calculated for each subject, topic, and week. These are shown in Table 1. The grand means for affective tone were 3.26 (SD = .83), 2.71 (SD = .81), and 2.38 (SD = .53) for mother, baby's father, and self, TABULAR DATA OMITTED respectively. The means of the ranges of ratings were 1.82 (SD = .94), 1.96 (SD = .92), and 1.89 (SD = .85) for mother, baby's father, and self, respectively.

The ratings of the affective tone of the three topics (averaged across raters and across weeks) were correlated to determine if they were similar in their orderings. The correlations were -.02 (mothers/selves), .01 (babies' fathers/selves), and -.18 (babies' fathers/mothers). None of the Pearson correlations were significant at the .05 level or high enough to discuss in terms of practical significance. This was interpreted as indicating a lack of relationship among the topics based on tonal ratings.

In order to have objective criteria with which to compare adolescents' feelings, a scatter plot was drawn with each adolescent's average tone on the horizontal axis and range on the vertical axis. The distribution was divided into nine sectors based on the observed clustering of data points. The neutral values for tone were between 2.5 and 3.5. The moderate values for range were between 1.5 and 2.5. Values below the neutral level were labeled as negative, and those above the neutral level were labeled as positive for tone. Values below and above the moderate levels for range were labeled low and high, respectively.

Table 2 shows how the participants were labeled based on their location in the nine sectors of the scatter plot. They felt more positive about their mothers than they did about themselves. Likewise, their feelings toward their mothers and babies' fathers were more variable than were their feelings toward themselves. Both the positive feelings toward their mothers and the neutral feelings toward their babies' fathers revealed high lability. Their self feelings were negative and moderately stable.

The following verbatim quotations from the participants' diaries are included to illustrate the findings reported above. Selected examples are given for each of the three topics.

Mothers

Participant 4 illustrated one of the two general patterns shown for feelings toward their mothers (positive tone/high range or neutral tone/moderate range). She was in the positive/high category because she felt positive about her mother (tone = 3.62), and yet was extremely labile in her feelings (range = 3.00). Participant 4 was 17 years old, was seven months pregnant, primiparous, single, resided in a residential facility for pregnant adolescents, and had parents with a high school education. The lability of her feelings can be seen in her descriptions of her mother as a "bitch," "very helpful," "unfair," "best friend," and "greatest mother in the world." The following are quotations from her diary which addressed the question: "During the last week I thought my mother was . . . because . . ."

Table 2
Distribution of Participants by Tone and Range of Feelings

 Participants By Numbers

Tone/Range Mothers Babies' Fathers Selves

Negative/Low 5,8 2 1,3,7,11
Neutral/Low -- 6 10
Positive/Low 10,11 -- --
Negative/Moderate -- 1,3,5 2,4,5,6,9,12,14
Neutral/Moderate 6,9,12,14 4,11 8
Positive/Moderate 3 14 --
Negative/High 1 10 --
Neutral/High -- 7,8,9,13 13
Positive/High 2,4,7,13 12 --

Note: Numbers represent the identification of participants

A bitch because she was not agreeing with me about marriage. She wouldn't let me get married and we fought about it. |week 1~

Very helpful because she backed me up on what I wanted and helped me feel wanted. She was there when I needed her and she stayed around even when I was unpleasant to be with. |week 2~

Sometimes she was unfair and sometimes she wasn't. We got along good for a while and then it went down the dumps. She bought me baby stuff and told me she wanted to meet my boyfriend. |week 3~

Like a best friend. She listened to me and understood what I was trying to say. |week 4~

The greatest mother in the world. I thought I went into labor friday and she took me to the doctor and stayed with me the whole time. When we were at her house she watched and made sure I was comfortable and feeling well. |week 5~

Table 2 shows that Participant 9 was included in the most frequent categories for each of the three topics (mother = neutral/moderate, father = neutral/high, self = negative/moderate.) Her feelings about her mother are described first. Her feelings about her baby's father and self are described under separate headings. She was neutral in her feelings about her mother (3.10), and these feelings were moderately variable (1.75). Participant 9 was 16 years old, married, experiencing her second pregnancy, and six months pregnant. Her parents were better educated than the other parents in the study. Her father had a bachelor's degree and her mother had a master's degree. Compared to Participant 4, Participant 9 was not as labile and intense in her feelings about her mother. She described her as "helpful," "lonely," and "nosey." "During the last week I thought my mother was . . . because . . ."

Helpful because we just got a new apartment and she came over and helped us move in. |week 2~

Lonely because I went to my grandma's for a family get-together and she kept hanging on me and telling me how much she missed me and that we should visit more. |week 4~

Nosey because she keeps asking is everything all right, are you okay and how is everything. I think she's waiting for me to come running back home and to say I was wrong. |week 5~

Babies' Fathers

Participant 8 was in the most frequent category of feelings toward the babies' fathers. She felt neutral about her baby's father (2.58) and was extremely labile (3.00). She was 17 years old, lived with her father and stepmother, was single, pregnant seven months, experiencing her second pregnancy, and had parents with a high school education. The high lability of her feelings toward her baby's father are reflected in her description of him as "very supportive," "mature," "very insensitive," "understanding," and "incredibly irresponsible." The following quotations show her feelings. "During the last week I thought my baby's father was . . . because . . ."

I haven't been able to see him as much as I like so I have been feeling neglected. But I know he needs to work. He's very supportive, I understand and I'm grateful to have him. He's a little irresponsible at times but I love him a lot. |week 1~

We have both matured a lot. Nine months ago he didn't have a job and was living with friends. Now he's holding two jobs and has his own place. Of course, he doesn't have a lot of time for us to see each other and sometimes I feel a little neglected and fear we are going to drift apart. And since we have decided to release the baby for adoption, I am scared that after the baby's gone there will be nothing to hold us together. I really hope that doesn't happen. |week 2~

He has been very insensitive lately. We've made plans and he always wants to be with his brother and friends and play. I'm not really fond of his friend so I don't like to be around with him when he is around. |week 3~

At the first of the week we broke it off. Sometimes I feel like I'm giving up so much of my life because of this. I wanted to do so much this year like drill team, prom, debate, and a lot of social activities I'm missing. His life hasn't changed at all. Then he seems like I have it so easy. Anyway we got back together Wednesday. We were only broke up for a couple of days, but it was really depressing. |week 4~

He's been a lot more understanding and caring so our relationship is doing a lot better now. I hope it stays this way. I just hope nothing happens to him with the law. If he went back to jail it would definitely make my delivery harder if he wasn't with me. |week 5~

He's incredibly irresponsible. We'll make plans and he doesn't show up so I have to hunt him down and that always makes me real mad and we end up in a fight. Tuesday we almost missed lamaze class because of that. |week 6~

Participant 9 also was neutral (3.16) about her baby's father, who was also her husband, and highly variable (3.00). She used words such as "supportive," "mean," "helpful," "excited," "lazy," and "irritating" to describe him. "During the last week I thought my baby's father was . . . because . . ."

Supportive. He usually is but during the last week he's been helping around the house. Also when I get angry and yell he doesn't get upset, he just ignores it and goes on. |week 1~

Mean because he takes away all of the food I like to eat and gives me plain food. |week 2~

Helpful. He helped with the work around the house so I wouldn't be so tired all the time. Also he was concerned with my health and the baby's health. |week 3~

Excited. He can't wait for the baby to be born so he can take it places and show it to people. I really think he will be a good father to the baby. |week 4~

Lazy because he didn't want to help get the housework done and usually he helps a lot. But this week our house has looked like a mess. |week 5~

Irritating because it seemed like everything he did made me want to scream and get upset and walk out. Hopefully it will pass after I have the baby. |week 6~

Selves

Participant 7 represented subjects who felt consistently negative about themselves. The tone of her self-description was 1.47 and her range was .58. She was 17 years old, single, lived with foster parents, three months pregnant, primiparous, and had parents with less than a high school education. She showed a fairly high level of stability in describing herself as "tired," "lazy," "bitchy," "rude," and "under a lot of stress." "During the last week I thought I was . . . because . . ."

Being pushed to the limits because my job wanted me to work everyday and I need to get my life organized but it is impossible. |week 1~

Always being tired and lazy because I felt like crap all week. I was sick all day. |week 2~

I was being a bitch to everyone because I was sick all of the time. |week 3~

I was being a bitch to everyone. I was in a bad mood all week and was rude to everyone at work and at home. |week 4~

Too tired because I've been thinking about my real dad who is in the hospital. |week 5~

Being under a lot of stress because my real dad was in the hospital and he treated me like crap when I spent $78 to go to Utah to see him. |week 6~

Participant 9 was negative in her feelings about herself (2.16) and moderately labile (2.50). She had a moderately labile score because, although there were some variations, her self-descriptions included "cranky," "lazy," and "grumpy." "During the last week I thought I was . . . because . . ."

Cranky because I haven't gotten enough sleep and any little thing can make me upset or irritated. I realize the way I am acting but I can't help the way I act. |week 1~

Lazy because moving made me tired so I was sitting around a lot which made me feel like a bum. |week 2~

Grumpy because I was tired from my vitamins not working so I guess I haven't been the easiest person to get along with. |week 3~

Growing up. I feel that I will be able to handle the responsibility of having a baby and I feel I will be a good mother and so do my family members and they are all excited too. |week 5~

CONCEPTUAL DISCUSSION

Although there is an established tradition of survey research with pregnant adolescents, there is a serious limitation in surveying them at one point in time. It is very difficult to obtain homogeneous samples in terms of age, gestational progress, and life characteristics. The heterogeneity of samples, coupled with pregnant adolescents' labile emotional states, and tendencies toward social desirability, make generalizability questionable. The topic of mood variability among pregnant adolescents deserves considerable attention for two related reasons, one practical and the other methodological. First, pregnant adolescents are forced to make critical decisions related to pregnancy outcome, marriage, employment, and family relationships at times when they also are experiencing emotional mood swings. Second, self-report data from pregnant adolescents that are used as the basis for public policy decisions, theory development, and service delivery models may be unreliable and of questionable validity. There is a need for much more research and theory on adolescent mood states during pregnancy that will help explain how the timing, intensity, and context of such moods interact with adolescent decision making.

The main purpose of this pilot study was to illustrate intrapregnancy variations in pregnant adolescents, an area without previous research or a theoretical foundation. The results are preliminary and not necessarily generalizable to a larger population of pregnant adolescents. The sample represented primarily low to moderate income white adolescents in the second trimester of pregnancy who were generally not living in intact two-parent families. Research will be needed with larger samples to determine if the patterns revealed here hold true for pregnant adolescents of other races and socioeconomic levels, as well as in different stages of pregnancy.

Causality cannot be determined from these findings if (1) events in their lives or hormonal changes influenced their moods; (2) there was an interaction between environmental and physiological factors, or (3) without prospective data if these moods were due to the state of being pregnant or more stable personality traits.

Nesselroade (1991), in discussing the measurement of change, makes a strong case for the measurement of three samples: persons, variables, and occasions. In the present study, the desire to sample six occasions and gather qualitatively rich data from each adolescent necessitated using a relatively small number of participants. Nesselroade (1991) stated that developmental research requires more rather than fewer occasions, other things being equal. He further stated that "it may be as important or more important to increase the number of occasions as the number of subjects when design constrains permit". To capitalize on the strengths associated with multiple occasions without influencing the biases associated with a narrow selection of subjects, he suggested investigating intraindividual variation and change patterns on multiple individuals. Nesselroade believes that the characteristics of intraindividual variability at the individual level, such as means, variances, and periodicities, can be the basis for studying both intraindividual change and stable interindividual differences. "Combining intraindividual variability and intraindividual change . . . leads to a representation of the developing individual that accents the variable and occasion sampling concerns". His recommendation was followed here in terms of investigating intraindividual variation in 14 different individuals. Intraindividual change was not within the scope of this study, but it is hoped that it will be addressed by future researchers.

In terms of lability, it is clear that these adolescents were very aware of their mood swings in relation to their mothers and babies' fathers. Several of them used the word "mood" in their diaries. Participant 13 described her moods in detail, as can be seen in the following selected excerpts from her diary. "During the last week I thought I was . . . because . . ."

Very emotional and lonely. I felt as if I was having second thoughts but after a while I would return to being happy, but like a pattern I would again turn to being depressed. |week 1~

Moody because everything that went wrong at all I cried over and when things would go good I'd be excited and five minutes later I'd be down again. |week 5~

Still really moody and a little depressed because my doctor changed my due date again by two weeks! I don't know how far along I am! |week 6~

In terms of contextuality and emotional tone, the results illustrated that different patterns of feelings were held for different individuals in the adolescents' lives. These patterns were positive and labile feelings toward their mothers, moderate and labile feelings toward their babies' fathers, and negative and stable feelings toward themselves. Their feelings on each of the topics were not interrelated; they were independent and context specific. Only two subjects (1 and 5) had similar ratings across all three topics, and these were all negative. On the other hand, six subjects (4, 7, 10, 11, 12, 14) had feelings that were labeled as positive (mothers), neutral (babies' fathers), and negative (selves) for the three topics.

Although in their self-descriptions several of the adolescents felt that they treated everyone around them negatively, they did not necessarily express negative feelings toward their mothers or babies' fathers. For example, Participant 7 was extremely negative in her self-ratings and yet she felt very positive about her mother and neutral about her partner. Of the four females who felt negatively about themselves, only one (Participant 1) felt negatively about her mother and one (Participant 3) felt negatively about her partner.

Future Research

A large number of questions need to be addressed in future research: Do mood swings during adolescence compound "normal" mood swings during pregnancy? How do mood swings of pregnant adolescents compare with those of pregnant adults? Will mood swings have worse consequences for the pregnancy during adolescence because of differences in levels of social support from others or because of differences in cognitive development? What is the relationship between mood variability and depressed mood state, and the potential consequences of each?

There is room for much creativity in terms of methodologies in exploring these questions. According to Watson and Tellegren (1985), positive and negative affectivity may be orthogonal; a person can be simultaneously high on both. Future researchers could explore two separate ratings of positive and negative emotional tone for the same individual at a given time. Another variation may be to ask raters to give global ratings of mood changes across time for each individual.

In addition, the participants themselves could be asked to complete instruments measuring dispositional optimism (see Scheier & Carver, 1985), or meta-mood (see Mayer & Gascheke, 1988). The Mayer and Gaschke Meta-Mood Experience Scale measures personal perceptions of one's overall mood experiences. For example, one question asks if the subject agrees that his or her mood is in agreement with the way the world is. A kind of multimethod comparison could be made between scores on these instruments and the assessments of affective tone provided by the raters of the diaries. If the same methodology is followed as was done here, other topics could be added to or substituted for mothers, babies' fathers, and selves, such as feelings toward their future, school, and babies. Finally, the individuals referred to in the diaries could be asked to complete their own diaries, similar in design to those completed by the pregnant adolescents, describing their feelings about the girls over time.

Proposed Hypotheses

Tentative hypotheses resulting from this study that need to be tested with larger samples include the following: First, pregnant adolescents' feelings about and relationship with their babies' fathers do not influence their feelings about their mothers, or vice versa. Second, pregnant adolescents feel more negatively about themselves than they do about their babies' fathers or their mothers. Third, their negative self-images are more stable than their feelings about their mothers or babies' fathers, thereby making self-image data more reliable than data on their feelings about the other two people.

Implications

If pregnant adolescents are generally as moody as those studied here, there are major implications for the reliability of self-report data so often gathered with this population. It is not sufficient, when studying pregnant adolescents, to report the reliability coefficients of the instruments used when these original values were based on nonpregnant and/or nonadolescent samples. New reliability coefficients need to be calculated and reported for the sample of pregnant adolescents currently of interest. If homogeneous samples are not available, at least the mean number of months pregnant and standard deviation need to be reported. Number of months pregnant also might be considered as an independent variable. Service-related and public policy decisions are often made based on one-time, self-report survey data from pregnant adolescents. These data must be viewed with a skeptical eye given the fact that the adolescents might have been experiencing emotional mood swings at the time of administration, thereby bringing into question the test-retest reliability of their scores.

The practical implications for this research are related to helping pregnant adolescents make effective decisions. Practitioners need to be aware of (a) the lability and specific patterns of the moods of each of the pregnant adolescents with whom they work, and (b) the context-specific nature of these moods. Critical decisions made during periods of elevated or depressed mood states may have negative and irrevocable implications in the adolescents' future. Pregnant adolescents need to be helped to chart their own personal mood patterns and to understand how these moods can affect their perceptions of the people and events in their lives, affecting their decision making.

REFERENCES

Ballinger, C. B., Kay, D., Naylor, G., & Smith, A. (1982). Some biochemical findings during pregnancy and after delivery in relation to mood change. Psychological Medicine, 12, 549-556.

Beck, A. T., Rush, A. J., Shaw, B. P., & Emery, G. (1979). Cognitive theory for depression. New York: Guilford.

Bence, P. J. (1990, March). Mood variability: Detecting cyclicities in the experience of mood during adolescence. Paper presented at the Biennial Meeting of the Society for Research on Adolescence, Atlanta, GA.

Bence, P. J. (1992, March). Patterns of the experience of mood. Paper presented at the Biennial Meeting of the Society for Research on Adolescence, Washington, D.C.

Blinn, L. M. (1988). Stability and change in pregnant adolescents' perceptions of their future marriage and family life. Family Perspective, 22, 255-268.

Bradburn, N. (1969). The structure of psychological well-being. Chicago: Aldine.

Brooks, Gunn, J., & Reiter, E. O. (1990). The role of pubertal processes. In S. S. Feldman, & G. Elliott (Eds.), At the threshold: The developing adolescent (pp. 16-54). Cambridge, MA: Harvard University Press.

Buchanan, C. M. (1991). Pubertal states in early adolescent girls: Relations of moods, energy, and restlessness. Journal of Early Adolescence, 11, 185-200.

Buchanan, T. (1991, April). Predicting change in postpartum depression: An individual growth curve analysis. Paper presented at the Biennial meeting of the Society for Research in Child Development, Seattle, WA.

Campbell, A. (1981). The sense of well-being in America: Recent patterns and trends. New York: McGraw-Hill.

Cervera, N. (1991). Unwed teenage pregnancy: Family relationships with the father of the baby. Families in society: The Journal of Contemporary Human Services, 72,(1), 29-36.

Crumidy, P. M., & Jacobziner, H. (1966). A study of young unmarried mothers who kept their babies. American Journal of Public Health, 56, 1242-1251.

Csikszentmihalyi, M., & Larson, R. (1987). Validity and reliability of the experience sampling method. Journal of Nervous and Mental Disease, 174, 526-536.

Diener, E., Sandvik, E., & Larsen, R. (1985). Age and sex effects for emotional intensity. Developmental Psychology, 21, 542-546.

Fitts, W. H. (1965). Tennessee Self-Concept Scale. Nashville: Counselor Recordings and Tests.

Friedman, H. L. (1966). The mother-daughter relationship: Its potential in treatment of young, unwed mothers. Social Casework, 47, 502-506.

Giblin, P., Poland, M. L., & Sams, B. A. (1987). Effects of social supports on attitudes and health behaviors of pregnant adolescents. Journal of Adolescent Health Care, 8, 273-279.

Larsen, R. J. (1987). The stability of mood variability: A spectral analytic approach to daily mood assessments. Journal of Personality and Social Psychology, 52, 1195-1204.

Larsen, R. J., & Kasimatis, M. (1990). Individual differences in entrainment of mood to weekly calendar. Journal of Personality and Social Psychology, 58(1), 164-171.

Larson, R., & Lampman-Petraitis, C. (1989). Daily emotional status associated with the onset of adolescence. Child Development, 60, 1250-1260.

Larson, R., Csikszentmihalyi, M., & Graef, R. (1980). Mood variability and the psychosocial adjustment of adolescents. Journal of Youth and Adolescence, 9, 469-490.

Mayer, J. D., & Gaschke, Y. (1988). The experience and meta-experience of mood. Journal of Personality and Social Psychology, 55, 102-111.

Nesselroade, J. R. (1991). Interindividual differences in intraindividual change. In L. A. Collins, & J. L. Horn (Eds.), Best methods for the analysis of change (pp. 92-106). Washington, DC: American Psychological Association.

Offer, D., Ostrov, E., & Howard, K. I. (1982). The Offer Self-Image Questionnaire for Adolescents: A Manual. Chicago: Michael Reese Hospital.

Panzarine, S. (1983). Adolescent mothers: Perceived stressors, social support, and coping during puerperium. Unpublished doctoral dissertation. University of Utah, Salt Lake City.

Perez, R. (1983). Effects of stress, social support and coping style on adjustment to pregnancy among Hispanic women. Hispanic Journal of Behavioral Sciences, 5, 141-161.

Piers, E. V., & Harris, D. B. (1983). The Piers-Harris Children's Self-Concept Scale. Los Angeles: Western Psychological Services.

Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press.

Salovey, P., & Birnbaum, D. (1989). Influence of mood on health-relevant cognitions. Journal of Personality and Social Psychology, 57, 539-551.

Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health: Assessment and implications of generalized outcome expectancies. Health Psychology, 4, 219-247.

Sharp, V. (1980). Adolescence. In J. Bemorad (Ed.), Child development in normality and psychopathology (pp. 174-218. New York: Brunner/Mazel).

Smith, E. W. (1975). The role of the grandmother in adolescent pregnancy and parenting. Journal of School Health, 45, 278-283.

Smith, R., Cubis, J., Brinsmead, M., & Lewin, T. (1990). Mood changes, obstetric experience and alterations in plasma cortisol, beta-endorphin and corticotrophin-releasing hormone during pregnancy and the puerperium. Journal of Psychosomatic Research, 34, (1), 53-69.

Spielberger, C., Gorsuch, R., & Lushene, R. (1970). Manual for the State-Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press.

Watson, D., & Tellegren, A. (1985). Toward a consensual structure of mood. Psychological Bulletin, 98, 219-235.

Young, A., Berkman, B., & Rehr, H. (1975). Parental influence on pregnant adolescents. Social Work, 20, 387-391.

Zekoski, E. M., O'Hara, M. W., & Wills, K. E. (1987). The effects of maternal mood on mother-infant interaction. Journal of Abnormal Child Psychology, 15, 361-378.

Lori Stenberg, Assistant Professor, School of Home Economics, University of Idaho, Moscow, Idaho 83843.

Cecelia Thompson, Associate Professor, College of Education, University of Arkansas, Fayetteville, Arkansas 72701.