Measurement is a necessary element of scientific research. The purpose for developing a new scale is the researcher’s belief that previous scales are insufficient, or they do not entirely cover the construct being studied. Scale development can be broken down into four phases—development, administration, evaluation, and finalization—consisting of eight steps. Phase 1 involves defining the construct to be measured (Step 1), generating an item pool (Step 2), having experts review the items (Step 3), modifying the item pool (Step 4), and formatting the draft scale based on expert feedback (Step 5). Phase 2 administers the draft scale to participants (Step 6). Phase 3 evaluates the responses for reliability and validity (Step 7). Phase 4 optimizes the scale’s length (Step 8). These eight steps of scale development will be demonstrated in this case through the recent development of the Myths & Attitudes About Fathers scale.
By the end of this case, students should be able to
- Understand the importance of scale use in research
- Recognize the eight steps of scale development
- Be inspired to think of new ways to establish research areas
Developing a scale is a significant undertaking as it is a research project in itself. Oftentimes researchers chose to develop a scale because there is not already one published to utilize in their research. The Mental Measurements Yearbook is a great tool to help one find out whether there currently is an instrument to use as the measurement portion of their research study.
I created the Myths & Attitudes About Fathers (MAAF) scale because at the time, there was not a scale to measure people’s attitudes and beliefs about fathers. There were measurements that looked at a father’s perception of himself and a child’s perception of his or her own father, but not one where people indicated their view of fatherhood. The MAAF scale assesses attitudes about fathers to compare the results with research. The scale applied a Likert scale to statements from research to assess the range of beliefs about fathers. Results that go against what research supports are considered a myth. Future studies will use the scale to assess different populations to see whether there is a common misconception that can be corrected.
I designed the scale for two main reasons. First, to measure attitudes about fathers with expectations to understand why nonresident fathers might not be involved in their child’s life, as statistics show a decline in involvement from time of separation for many fathers. Second, I was looking for a reason as to why in the United States there is a stereotype that mothers are more important.
The role of fatherhood in U.S. society has changed over the years. Hundreds of years ago, the father was the most important parent for raising the children, then he became the breadwinner, and now there is an expansive volume of research on the lack of father involvement in child-rearing. Plenty of fathers want to be a part of their child’s life and do whatever they can to do so. However, many fathers encounter barriers created by possible myths which limit, or in some cases prevent, their ability to engage with their children. Commons myths that people might believe about fathers include the following:
- Fathers are not interested in being involved;
- Fathers do not have the capability to be involved;
- Fathers are harmful if they are involved;
- There is little to no effect if the father is not involved (or the hassle of dealing with the other parent is worse than negative effects on children).
Fathers’ determination to assume their rightful role in their child’s life is not enough. Both parents are important—not just to feed, bath, and shelter children. Caretaker importance goes beyond taking care of the physical and safety necessities of infants and children.
Scales are measurement instruments comprising an assortment of items combined into a composite score and are developed to measure phenomena that are thought to exist because of theoretical understanding but cannot be assessed directly (DeVellis, 2012). Developing a valid scale is a difficult and time-consuming progression. However, the results are more precise and reliable in assessing the underlying construct. Scales have four advantages over single-item measures: enhanced validity, enhanced reliability, better level of measurement, and better efficiency in data handling (Monette, Sullivan, & DeJong, 2011).
The development of a new scale presumes that the researcher has clearly determined a construct and adequately reviewed the literature to conclude that an appropriate scale does not already exist or help determine whether a scale is needed at all (Clark & Watson, 1995). The review should establish the scope and identify the most important information needed to begin the necessary content of the scale. The review can also reveal useable questions from other selected instruments (McKenzie, Wood, Kotechi, Clark, & Brey, 1999). The literature review may also help to recognize complications with existing measures, such as vague instructions or debatable response formats, that can be evaded in the development of a new scale (Clark & Watson, 1995)
In scale development, a critical first step is having a well-defined construct, and many developers consider it the most challenging step (McKenzie et al., 1999). Accurate scale content requires thinking clearly about the construct being measured (DeVellis, 2012). The creation of the construct and its objectives provide the developer with an outline of what should be included in the scale (McKenzie et al., 1999). Generating the initial list of questions (item pool) is considered one of the most significant parts of developing a scale (Clark & Watson, 1995). The constructs established in Step 1 are used as a classification of paradigms. The definitions are used to guide the development of the item pool (Schwab, 1980).
Prior to administering a scale to a sample, it can be given to a panel of experts for their outlook and response on items and format. They can confirm or invalidate the construct by rating how relevant they think each item is to what is intend to measure. In addition, they can evaluate the item’s clarity and conciseness. Moreover, they can identify other means of including phenomenon that the researcher may have missed (DeVellis, 2012).
The original idea for this project was to assess myths that U.S. society believes regarding nonresident fathers (i.e., fathers who do not live with their child). Fatherhood myths are found in popular literature, movies, and other media. There had been a bias toward men in the U.S. court system (family court is handled at the state level and laws differ from state to state). My hypothesis is that there are myths about fathers and their capabilities and interests in being an involved parent (especially nonresident parent) and that not having a father in one’s life does have a large impact on children. After an in-depth review of the literature, I was not able to find a current measurement to assess attitudes people held about nonresident fathers or fathers in general. My goal was to create a scale that could identify whether or not people actually endorse fatherhood myths by agreeing or disagreeing with supported research.
There is a great deal of research supporting the importance of nonresident father involvement, but yet approximately 30% of U.S. nonresident fathers are not consistently involved in their children’s lives and 10% see their children less than once a month. This raises the questions of “why”? Research supports that some of the reasons are lack of prenatal involvement, relationship problems, inability to pay child support, socioeconomic status, and characteristics of the child, but is there more to it than that? Do people tend to believe myths about nonresident fathers due to images portrayed in media instead of believing what has been found in research? My research hypothesis is that yes, society believes more in myths (not supported by research) than in what research supports about nonresident father involvement.
To assess attitudes regarding nonresident fathers, I looked into numerous scales to see whether one would adequately measure how a person views fatherhood. I found some scales regarding fathers; however, none measured attitudes about fatherhood. The Perceptions of Parental Role Responsibilities (Gilbert & Hanson, 1982) was too generic, asking questions about whether or not it was the parent’s responsibility to teach the child certain activities. The Fatherhood Scale (Dick, 2004) measured adult male’s perceptions of his father while he was growing up. The Ackerman–Schoendorf (Ackerman & Schoendorf, 1992) questioned the parents of the child to see which parent was more fit in a custody situation. The Parental Attitude Research Instrument (Schaefer & Bell, 1958) and The Parent–Child Relations Questionnaire (Roe & Siegelman, 1963) surveyed only the parents, measuring parents’ behavior toward their children. I also looked at the Attitude Toward Men Scale (Maltby & Day, 2001), but it contained only six questions regarding fatherhood.
Although none of these scales measure attitudes regarding nonresident fatherhood, they did offer some suggestions of types of questions to use in the development of a scale. In addition to these scales, several articles also helped me in developing the MAAF scale, including those by Jessica Troilo and Marilyn Coleman (2008); Ruth Chu-Lien Chao and Kathy E. Green (2011); Donna Brown, Scott Ryan, and Janet Therese Pushkal (2007); and Benjamin E. Caldwell and Scott R. Woolley (2008).
I based the construction of the MAAF scale on Robert F. DeVellis’s (1991) eight-step model of scale development. The steps include the following:
- Determine clearly what you want to measure;
- Generate an item pool;
- Determine the format of the measure;
- Expert review;
- Consider inclusion validation items;
- Administer items to a development sample;
- Evaluate the items;
- Optimize the scale’s length.
I followed these steps and conducted two studies: Study 1 focused on the development and expert review of the scale by defining the construct to be measured, generating an item pool, having experts review the items, modifying the item pool, and formatting the draft scale based on expert feedback. Study 2 administered the draft scale to participants, evaluated the responses for reliability and validity, and optimized the scale’s length.
As a validation measure, it is wise to consider including additional items that can help in determining the validity of the final scale. A pool of items is not automatically a scale. The items need to be administered and examined to see how well the results confirm expectations about the scales structure (Hinkin, 1995). Next to item development, item evaluation is second in the level of importance for scale development (DeVellis, 2012).
The first phase of Study 1 focused on the initial development of potential items. Based on information from the literature review, this phase helped me to determine clearly what to measure. I generated an item pool based on the constructs from the literature, and then choose the format of the measure.
Based on a literature review, I chose four areas as the focus for the MAAF scale: father involvement, type of custody, attachment to father, and potential risks. Each of the four constructs gleaned from the literature were then operationalized as follows:
Type of custody
Attachment to father
- Father involvement. Father involvement was described by several variables: financial support, frequency and length of time spent, engagement, accessibility, and responsibility (Hook & Chalasani, 2008). Engagement incorporated all methods of direct contact among fathers and children, such as care and play (Hook & Chalasani, 2008). Accessibility included the father’s availability to children (Hook & Chalasani, 2008), regardless of the nature or degree of relations between father and child (Lamb, Pleck, Charnov, & Levine, 1985). Responsibility included understanding and meeting the needs of the child by providing economic resources to the child and planning for the child’s future (Lamb et al., 1985).
- Type of custody. Type of custody was described as the right of determining the residence, protection, care, and education of a minor child or children, in a divorce or separation (Bauserman, 2002). There are three basic types of custody in the United States: sole, joint physical, and joint legal. Sole custody means that only one parent is able to legally make all decisions regarding the child (Pasley & Braver, 2004). Joint custody means both parents share the responsibilities of parenting. The two types of joint custody include physical/residential joint custody which is considered ongoing physical contact by both parents and joint legal custody infers shared decision-making by the parents and ongoing, active involvement of the nonresident parent, even if residential custody remains mainly with one parent (Bauserman, 2002).
- Attachment to father. Based on attachment theory, attachment to father was described as an intense feeling of intimacy, emotional security, and physical safety (Peluso, Peluso, White, & Kern, 2004) and child–father attachment as children needing both a secure base and someone they can explore with. Secure attachment was categorized by passionate feelings of intimacy, emotional security, and physical safety in the company of an attachment figure (Ainsworth, Blehar, Waters, & Wall, 1978).
- Potential risks. Risk elements are variables that reduce the likelihood of achieving personal goals or areas that society views as important. Potential risks were defined as an exposure to loss or harm, a hazard, danger, or peril (Dudley & Stone, 2001). Potential risks incorporated conditions that impeded a consistent pattern of father participation with his children (Fagan & Palkovitz, 2007). Risk is associated with psychological or social factors that intensify the probability that an individual will experience poor outcome in behavior, in health, academic achievement, or in regard to vocational success (Fagan & Palkovitz, 2007).
The second step involved generating a large pool of items for use in creating the final scale (DeVellis, 2012). In total, 108 items were generated from literature, the Perceptions of Parental Role Responsibility scale, and the Fatherhood Scale: 78 from the literature review, 12 items reworded from the Fatherhood Scale, and 18 items reworded from the Perceptions of Parental Role Responsibility.
I chose an online survey format for administering the item pool for several reasons: ease of administration, ability to solicit a more diverse sample pool, and cost-effectiveness. I administered the scale online through a SurveyMonkey link and a Facebook account linked to SurveyMonkey. Participants received an electronic consent form, a demographics questionnaire, three informational questions about the participant’s relationship with their father and their view of parenting, five social desirability questions, the Adult Attachment Scale (optional) (Collins & Read, 1990), and the MAAF item pool. The item pool was presented as a 6-point Likert-type response (1—disagree very much to 6—agree very much).
Participants were needed for each of the two studies. The first study required participants with expertise in statistics, scale development, and/or families to aid in the development of the draft scale based off of the item pool. The second study required participants from the general population. Their responses aided me in the evaluation of the draft scale to retain items for the final MAAF scale.
I chose 12 experts (three scale developers, two statisticians, six counselors, and one psychologist) based on a set of criteria which included occupation, experience, knowledge, and availability. I emailed each expert individually, and six responded (5—yes, 1—no). The experts included one statistician, one psychologist, one scale developer, and two counselors. The experts who agreed to participate received a second email thanking them for their help, the link to the item pool and an attachment explaining the directions and constructs.
The actual expert review was only completed by four experts (one statistician, one psychologist, and two counselors). Even though the expert review did not go as planned, it still helped me in the construction of the scale. In the event of future scale construction, I would still perform this optional step.
The scale was open to anyone more than the age of 18 years. The MAAF draft scale had a sample size of 332 initial participants who completed the demographics (Table 1). The number decreased to 326 for the additional information questions (Table 2) and the social desirability questions (Table 3). In all, 269 participants completed the MAAF scale and 261 participants completed it along with the Adult Attachment Scale. In SPSS, a software package for statistical analysis, if a person skipped one or more of the questions, their data are deleted from the analysis, so only 212 respondents were calculated. The case processing summary can be found in Table 4.
|Table 1. Demographics.|
Age 70 or older
American Indian or Alaskan Native
Black or African American
Income US$200,000 and up
Highest level of education: high school
Highest level of education: some college
Highest level of education: certification
Highest level of education: associates
Highest level of education: bachelors
Highest level of education: masters
Highest level of education: doctorate
Enrolled in school
Not enrolled in school
Other sexual orientation
In a committed relationship
|Table 2. Informational questions.|
10. What was your overall relationship like with your father as a child (0-12 years old)
Very involved (spent quality time with each other 5+ times a month)
Involved (spent quality time with each other 1-4 times a month)
Semi-involved (spent quality time with each other every 3 months)
Occasionally involved (spent quality time with each other 1-2 times a year)
Not involved (never knew if or when I would see or talk to him)
11. What was your overall relationship like with your father as an adolescent (13-18 years old)
Very involved (spent quality time with each other 5+ times a month)
Involved (spent quality time with each other 1-4 times a month)
Semi-involved (spent quality time with each other every 3 months)
Occasionally involved (spent quality time with each other 1-2 times a year)
Not involved (never knew if or when I would see or talk to him)
12. Which of these is a description of the way parenting should work?
Authoritarian parenting (children are expected to follow the strict rules established by the parents without any debate or explanation. Punishment is the result)
Authoritative parenting (establish rules and guidelines that their children are expected to follow. Rules are explained and discussed. Failure to follow rules results in more nurturing and forgiving rather than punishing)
Permissive parenting (parents have very few demands to make of their children. These parents rarely discipline their children, often taking on the status of a friend more than that of a parent)
Uninvolved parenting (few demands, low responsiveness, and little communication. Although these parents fulfill the child’s basic needs, they are generally detached from their child’s life)
|Table 3. Social desirability.|
13. I have never intensely disliked anyone
14. I’m always willing to admit it when I make a mistake
15. I always try to practice what I preach
16. When I don’t know something I don’t at all mind admitting it
17. I sometimes think when people have a misfortune they only got what they deserved
|Table 4. Case processing summary.|
The first phase of Study 2 focused on the inclusion of validation items: social desirability items and construct validity items. To assess social desirability, I added five items from the Marlowe–Crowne Social Desirability Scale (MCSDS) to the list of questions administered before participants actually answered scale items. Construct validity was measured using reworded items from the Perceptions of Parental Role Responsibility and the Fatherhood Scale (as discussed in “Generate an item pool” section). Also, I used the Adult Attachment Scale as a co-construct to help with construct validity as a discriminant function. I gave participants the option of completing this scale or not completing it. Of the 212 completed surveys, 204 participants completed the Adult Attachment Scale.
The second phase of Study 2 focused on administering items. I offered the MAAF draft scale in an online format and provided the consent form to participants electronically. Once the consent was electronically accepted, the participants received a demographic questionnaire, additional information questions, social desirability questions, and a draft of the MAAF containing reworded questions from the Fatherhood Scale and the Perceptions of Parental Role Responsibility.
The MAAF scale included items from three additional scales. The Perceptions of Parental Role Responsibility scale measures parent’s attitudes about their children. The Fatherhood Scale measures adult children’s attitudes about their father while growing up. I reworded items from Perceptions of Parental Role Responsibility and the Fatherhood Scale for fathers in general instead of a father’s view of himself or the adult child’s view of their father. (I obtained permission from both authors.) The Perceptions of Parental Role Responsibility scale has a coefficient alpha ranging from .81 to .91, with an average of .86 and has a 25% to 30% variance associated with any two particular scales shared, thus indicating that each scale represents an important and relatively independent dimension of perceived parental role responsibility. The Fatherhood Scale has an overall reliability of 0.98 and a factor analysis produced 13 factors that accounted for 75% of the total variance. Also, the Adult Attachment Scale was used as a co-construct to help with construct validity as a discriminant function. Participants were given the option of completing this scale (n = 204).
Data analysis focused on evaluating the results from the draft scale. I administered the items to 330 people, of whom 212 answered all of the questions (n = 212) and calculated a factor analysis and coefficient alpha to see how well the results confirmed expectations about the scale’s structure (Hinkin, 1995). Overall evaluation covered (a) initial item analyses via exploratory factor analysis, (b) initial item analyses and internal consistency estimates, and (c) initial estimates of validity.
I then reviewed the MAAF scale for missing items, improperly marked items, personal identifiers, evidence of confusion, and items that have exceedingly skewed and uneven distributions. Next, I ran descriptive statistics (mean and standard deviation) for all of the MAAF items based on 212 responses. No items needed to be removed or corrected. However, I noted that more detailed instructions should be given at the beginning of the scale because I received feedback that there were situations that altered a participant’s answer. In future studies, I will include directions that state that the participant should rank the items based on an overall opinion of fathers.
I conducted factor analysis using SPSS 21.0. Running the Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy and Barlett’s test of sphericity determined that the MAAF scale’s items were similar enough to run an exploratory factor analysis, so I ran an orthogonal factor analysis with varimax, measuring the total amount of variation observed in all variables that had been selected. After extracting the factors, I conducted eigenvalues. Using the Kaiser rule (the default in SPSS), I dropped components with eigenvalues below 1.0.
A total of 17 factors were extracted, which is a large amount. Kaiser’s criterion is not recommended as the sole determination for approximating the number of components because it often over-extracts items. Components with eigenvalues of 1.00 to 1.9 account for only 1% of the information within a scale and are often considered too generous (DeVellis, 2012). Components 6 to 17 have eigenvalues that fall into that category resulting in five main components. A scree plot also suggested that five components be used. This agreed with my original theory that nonresident fatherhood consisted of four areas of assessment. However, it broke down the area of potential risk into two components, thus resulting in five components.
I then re-ran an exploratory factor analysis to confirm the 5-factor model. Components 1 to 5 account for 44.67% of the MAAF scale. Component 1 was labeled as father involvement, Component 2 as type of custody, Component 3 as attachment to father, Component 4 as barriers fathers encounter, and Component 5 as potential risks. Four of the five components matched the original constructs used for the development of the initial item pool (father involvement, type of custody, attachment to father, and potential risks). This confirmed my first hypothesis that an exploratory factor analysis of the MAAF would produce unique factors relevant to the literature.
To determine internal consistency (reliability), I ran coefficient alpha and item-total statistics for the MAAF scale. A score .94 was obtained, indicating that the items were similar and thus measuring the same construct. This confirmed my second hypothesis: Cronbach’s coefficient alpha for the MAAF scale would be between .70 and .95. I then ran item-total statistics to see how each statement affected the overall alpha level. There were no significant improvements based on the removal of any certain item.
The participant’s responses for each scale were averaged and then used to run a Spearman correlation (Table 5). The MAAF scale had one item that needed reverse scoring, the Fatherhood Scale had four items, the Perceptions of Parental Role Responsibility had six items, and the Adult Attachment Scale had 11 items.
|Table 5. Correlations with the MAAF scale.|
MAAF: Myths and Attitudes About Fathers Scale; FS: Fatherhood Scale; PPRS: Perceptions of Parental Role Responsibility; AAS: Adult Attachment Scale. **Correlation is significant at the .01 level (one-tailed). *Correlation is significant at the .05 level (one-tailed).
Reducing the number of items within the draft scale is a very common and anticipated part of the process (Worthington & Whittaker, 2006). I retained items based on their loading score, face and content validity, item-to-total correlations, average inter-item correlations, and item variances (Netemeyer, Bearden, & Sharma, 2003). I deleted items that had loading scores above .80 and below .40, along with the items that contributed the least to the internal consistency and had low conceptual consistency with other items on the factor.
The final step was to optimize the MAAF scale’s length. Factor-analytic, data-reduction techniques produced a steady set of underlying factors that correctly mirrored the construct. My goal was to have a reliable scale that is concise. The MAAF scale’s item-scale correlations were equivalent to the average inter-item correlation. I then determined the number of items retained and the length of the MAAF scale based on the results calculated. The final version of the MAAF scale can be found in Appendix at the conclusion of this case study.
There were several limitations to this study. First, during data collection, a few participants noted that the wording of the items could be interpreted differently depending on the situation. In future studies, I will clarify the directions by asking the participants to rate each statement based on their overall opinion of fathers.
The second limitation was using a Likert scale. The limitation of using a Likert scale is that the distances between the descriptors can appear to be equidistant, but that the actual experiential distances appear to be more dissimilar (Lishner, Cooter, & Zald, 2008). I was able to reduce this by labeling only the ends of the distribution.
The third limitation was the size of the participant pool (266 recorded, only 212 analyzed). Although it is suggested that a sample sizes of 100 to 200 is acceptable for developing a scale (Hinkin, 1995), it can also be argued that there should be 5 to 10 respondents per item (Tinsley & Tinsley, 1987).
The fourth limitation was that there was only one strong component followed by several less strong components, thus indicating that the scale measured mainly father involvement.
The use of scales and its results should be approached cautiously. The amount of items on the scale needs to adequately represent a full variety of answers to avoid any type of bias. The items need to be presented in a neutral and nonjudgmental context (Monette et al., 2011). The different characteristics of a variable need to be well known and analyzed carefully because one may correlate with an independent or a dependent variable, whereas another may not. When viewing the results, it is important to make sure adequate development of the scale has taken place, especially in terms of reliability and validity.
Perhaps, the biggest limitation is that people may not make decisions about the importance of fathers based on myths. Perhaps, personal feelings and experiences are more predictive about decisions regarding father involvement. For example, a mother may not deny her former partner access to the child because she does not think fathers are important, but rather because she thinks her ex- is a “jerk.”
The ethical concerns of a scale are minimal as no treatment is being administered. It is a process of reducing an item pool by having an expert review the items and then administering a draft scale so that the final scale is valid and reliable.
For this study, participation was completely anonymous. Participants received an informed consent that covered the ethical concerns of the study. The benefits for participating in the development of the MAAF included feeling empowered and valued as part of assessing beliefs toward fathers. Participants may feel encouraged by having a chance to include their own voice in a research project which directly relates to the greater good of society.
Results of this study are intended to contribute to an action plan to educate society about the importance of father involvement, specifically in the areas where myths are believed. Although there are minimal risks for participating in this pilot study, one could experience feelings of discomfort or uneasiness in answering questions related to negative experiences with their father or someone else’s father. Participants may find that answering questions for this scale may reveal feelings to themselves that they were not previously aware of. Participants for this study were informed that if they felt uncomfortable or distressed at any time during this scale, they should feel free to terminate participation.
Developing a scale can be seen as a daunting project, but it is really not hard. I would encourage anyone who has a topic they want to research but for which no scale currently exists to create one. DeVellis’s eight-step model is straightforward and to the point. The lessons that I learned include doing a throughout literature review first. This helps in creating the item pool. Using current literature to create items provides a research basis for the results. I also advise using an expert review even though this step is considered optional. I liked doing this to identify unclear items and delete or revise them. The biggest lesson I learned occurred during the administration phase. SurveyMonkey lists the number of people who started the scale; however, there were several people who did only part of the evaluation thus making the numbers lower than expected. This resulted in a lower sample than what I thought I had. One way to help with this would be to make all questions required before someone could submit.
- Why is it important to choose questions rooted in research already published?
- Discuss the pros and cons of the optional expert review step. Based on your findings, would you include this step? Why or why not?
- Within social research, people often answer questions based on how they want to be viewed. Discuss the pros and cons of using social desirability questions within an instrument. Would you incorporate them? Why or why not?
- What makes a scale different from a questionnaire? In what types of studies would you suggest using a scale rather than a questionnaire? Why?
- What role does theory play in scale development?
- Attachment between a father and a child are essential to the survival of humans.
- Children who grow up lacking father involvement are worse off, on average, than children who grow up living with both of their biological parents.
- Lack of father input in a child’s life negatively affects school achievement.
- Children develop best when they are provided the opportunity to have continuous and enduring relationship with their fathers.
- Joint custody tends to promote more father contact.
- Fathers with joint custody spend approximately twice as many days with their children as fathers without joint custody.
- Joint legal custody increases nonresidential father involvement.
- Father’s with joint legal custody have closer father–child relationships.
- Joint custody is in the best of interest of children.
- Children and adolescents living in joint physical custody families are moderately better adjusted than children living with a sole custodial parent.
- Joint custody does not damage a child’s well-being.
- Positive relationship between parents makes it easier for fathers to be involved.
- Children’s experiences with their father affects their attachment styles.
- A mother’s attitude influences father–child relations.
- Custodial mothers are able to make choices about when and how the father spends time with his child.
- Nonresident fathers must manage a variety of obstacles within the court system.
- Judges prefer to give custody to mothers more often than to fathers.
- Social interaction between father and child creates feelings of attachment to the father.
- Father-child play is critical for child development.
- The impact of father involvement is more observable as the child grows older.
- Healthy development depends on a positive attachment to both parents.
- One of the problems of fatherlessness is a financial one.
- Children with separated parents are worse off because they do not have the money to give them a better life.
- Children close to their fathers are less distressed.
- Fathers have a matchless role to play in child development.
- Fathers are important to a child’s development.
- Growing up without father involvement harms children because they have less money to help aid in their child’s upbringing and education
- Growing up without father involvement harms children because parents have less time to spend with them.
- Growing up without father involvement harms children because it reduces access to community resources that can supplement and support the parent’s efforts.
- Children who lack father involvement are more likely to have lower self-esteem.
- Children who lack father involvement are more likely to score lower on measures of academic achievement.
- Adolescents who lack father involvement are more likely to drop out of high school.
- Marital uncertainty is higher for adults who experienced a lack of father involvement.
- Public policy creates a barrier for never married fathers.
- Joint residential custody (the child spends almost equal time with both parents) is the best arrangement for separated families.
- Father’s responses to the child shape the pattern of the child–father attachment relationship.