Introducing the Power In Nepali Girls (PING) empowerment scale

ABSTRACT

The multi-dimensional concept of empowerment is context specific, uniquely defined in various cultures, and challenging to measure. The aim of this study was to develop a scale for measuring empowerment among adolescent girls in Nepal. Twenty-nine items related to empowerment were drawn from formative, participatory research conducted in southern Nepal to comprise a draft empowerment scale. A case/control survey among 300 adolescent Nepali girls was then conducted to refine the scale. Factor analysis determined the most parsimonious scale and identified the underlying structure of items, resulting in the The Power In Nepali Girls (PING), a two-factor, 16-item empowerment scale. Factor 1 items address decision-making power and treatment in family and society. Factor 2 items address knowledge, education, and skills development. We used the 16-item scale to assess the impact of a social and financial skills program on empowerment among adolescent girls residing in southern Nepal. The PING scale is a culturally- and population-specific measure that can be used to quantify the impact of empowerment programs in Nepal among adolescent girls. The methods used also serve as a template for future work aiming to design context-specific measures of empowerment using community-engaged approaches.

KEYWORDS:

• Empowerment
• scale development
• adolescents
• Nepal
• factor analysis

Introduction

Global health, education, and development interventions addressing the United Nations Sustainable Development Goal of gender equality and empowerment are on the rise and are critical for improving the lives of women and girls (World Health Organization, 2021). Throughout the world, social norms, cultural expectations, and discrimination impact gender dynamics, the abilities of women and girls to make independent decisions, and their individual agency related to a wide range of issues including psychological, social and economic health and wellbeing. Growing evidence illustrates that empowerment is constructed from several conceptual elements including psychological, social, economic, legal, and political components (Pratley, 2016). While global action and attention to this issue has been amplified in recent years, the multi-dimensional concept of empowerment remains challenging to measure quantitatively and consistently, and as a result, evaluation evidence from empowerment programs remains largely qualitative and anecdotal.

Cyril et al.’s (2016) review of empowerment measures in global health promotion found that of the 20 relevant articles, a majority focused on research conducted in the United States, evaluated empowerment among adults, and focused on psychological empowerment (e.g. motivation and confidence) rather than decision-making and control. They conclude that ‘the [empowerment] instruments … have been developed using a western, individualist orientation towards empowerment’. Moving toward a more community-centred approach, The Bill & Melinda Gates Foundation Gender Equality Toolbox (Undated) defines empowerment as ‘the expansion of choice and strengthening of voice through the transformation of power relations so women and girls have more control over their lives and future’. The toolbox includes five guiding principles to measure empowerment: (1) tailor method to purpose and context, (2) take a holistic approach to measurement, (3) be informed by women and girls’ voices, (4) consider how gender intersects with age, class, ethnicity, and other social markers, and (5) understand that empowerment takes time. There is growing agreement that measures of empowerment should be context and setting specific (Cattaneo & Goodman, 2015; Zimmerman & Warschausky, 1998).

In Nepal, empowerment interventions are increasingly being used to address the numerous challenges that adolescent girls encounter, including the lack of awareness and knowledge about sexual and reproductive health, menstrual restrictions, child marriage, poor nutrition, school dropout, unsafe abortion, sexually transmitted diseases (STDs), HIV/AIDS, and substance abuse (Tamang et al., 2017; UNFPA, 2019; Chalise et al., 2018; Baumann et al., 2021; Government of Nepal, 2000). Though the government of Nepal has recognised adolescents as an under-served and vulnerable population with specific sexual and reproductive health needs, and has developed the National Adolescent Sexual and Reproductive Health Program, effective approaches for rigorously evaluating the impact of such interventions on youth empowerment are lacking (Government of Nepal, 2021).

Given that there is no known measure for assessing empowerment among adolescent girls in the context of Nepal, developing a Nepal-specific measure of empowerment is a critical next step for consistent evaluation of empowerment initiatives in the country. This article presents results from a two-phased study exploring empowerment among adolescent girls in Nepal. Phase 1 employed concept mapping, a participatory research method, to explore concepts of empowerment with 113 participants including national experts, program staff, adolescents, and their mothers. Participants first responded to the prompt: ‘The life of an adolescent girl improves when she has/can … ’ and then sorted and rated the resulting 105 items based on importance and clarity. Those same 105 items were grouped by the participants into four concepts: (1) Education & Knowledge, (2) Decision Making, (3) Supports & Skills and (4) Physical Infrastructure. During interpretation sessions the participants were asked about how the items identified as very important aid in empowerment to improve the lives of girls and how to refine the items to improve clarity and understanding. Additional details regarding the Phase 1 process and findings are presented in the associated 2020 UNICEF report. Phase 2 methods and results presented here led to the development of a Nepal-specific empowerment scale that was utilized to assess the impact of a holistic, social, and financial skills program on empowerment outcomes of adolescent girls residing in southern Nepal.

Methods

This project was conducted as part of a practice-research partnership with a long history of collaboration on sexual and reproductive health among women in Nepal. Funding from The UNICEF Regional Office for South Asia supported the community engagement and data collection efforts of practitioners from Nepal Fertility Care Center, a local, non-governmental organisation, and researchers from University of Pittsburgh contributed their time and effort to the project. The research activities were reviewed and approved by the Nepal Heath Research Council (748-2019) and the University of Pittsburgh Institutional Review Board (STUDY19090269).

Rupantaran program, partnership history, study development & design

UNICEF’s Rupantaran social and financial skills program focuses building adolescent girl’s knowledge and skills, such as enhanced literacy and numeracy skills. The program is designed to empower out of school adolescent girls with the knowledge and skills needed to make important life decisions. Since 2014, the Nepal Fertility Care Center has been implementing the program in Saptari, Rautahat and Dhanusha districts, three southern districts of Nepal.

The 15-module program aims to empower adolescent girls to participate in decisions affecting their lives and become change agents in their communities. The intervention facilitators, females with at least a 12th grade education and some teaching experience, are identified by the Nepal Fertility Care Center, in collaboration with the local government in the intervention regions. Facilitator training takes place over an intensive month-long, two-phase process. Intervention participants are underserved adolescent females, and are identified by the facilitators in collaboration with the local government in the intervention regions.

The intervention sessions are conducted with out of school adolescent girls and their parents. A total of 15 parent sessions are held concurrently alongside 80 adolescent sessions. The sessions average one hour in length, occur multiple times a week and are held at sites identified as convenient for the facilitators and participants. There is no financial incentive for participation. The sessions include information on a range of topics including self-realization, rights and responsibilities, nutrition, financial planning, sexual and reproductive health, and violence. The Rupantaran training also provides information on health and hygiene and entrepreneurship for improving livelihoods. The retention rate is high; most of the adolescent females attend all 80 sessions.

In 2019, the Nepal Fertility Care Center invited University of Pittsburgh researchers on a field trip to Southern Nepal to learn first-hand about the Rupantaran program, context, and impact. During that trip, the researchers concluded that while there are numerous success stories about the achievements of participants in the Rupantaran program, the impact of the intervention program on empowerment outcomes for the adolescent female participants had yet to be measured consistently. It was this lack of clarity and the fact that there is no known empowerment measure for the context of Nepal that led the team to develop a Nepal-specific measure of empowerment for adolescent girls using a two phased study design.

Participants

The sample addressed in this article includes 300 adolescent girls between the ages of 13–16 years old recruited from three districts in Southern Nepal, Province 2, specifically Saptari, Rautahat, and Dhanusha. One hundred and fifty girls were surveyed who completed the Rupantaran program (i.e. cases), and 150 were surveyed from matched communities that had not been exposed to the Rupantaran program (i.e. controls), specifically 50 cases and 50 controls from each of the three study districts. This sample size is based on standards outlined in the literature by Comrey (1973) for exploratory and confirmatory factor analysis (FA) studies for scale development, where a sample of 300 is considered appropriate for exploratory FA.

To recruit the case participants, a listing of the potential participants was made in each district with the support of Nepal Fertility Care Center field staff. Participants were eligible if they had completed the program within the past year and were between the ages of 13–16 years old at the time of data collection. Based on the required sample size, every third participant from the list was selected and contacted in-person to assess their interest in participation. The control group participants were selected by purposive sampling, with the goal of matching participants based on demographic characteristics (e.g., education, income, caste/ethnicity, religion, geographic region). To avoid contamination effects, the non-Rupantaran program participants were recruited from villages nearby the Rupantaran program sites, but not from the same village. The data collectors visited the villages and approached female youth of the same age as those completing the Rupantaran program to assess their eligibility and interest in study participation. Both case and control participants provided parental consent and youth asset. No compensation was provided for their participation in the study.

Data collection

To ensure objectivity of the survey data, a third-party agency was sub-contracted based on their research expertise, ability to speak the the local languages of the district, and their commitment to meeting the study timeline. The University of Pittsburgh researchers provided an in-person training to the agency for implementing a community-based, cross-sectional survey. The survey was administered via tablet by trained enumerators in the local language (i.e. Nepali or Maithili).

We assessed empowerment using the 29 items identified and refined during Phase 1 of this project (Table 2). The female adolescent participants were presented with items addressing issues including how their parents treat them with respect to boys, their ability to freely share and negotiate ideas, their access to transportation, and their knowledge of health-related issues. Each item was followed by a Likert-type response scale with possible responses of Strongly Disagree (1), Slightly Disagree (2), Slightly Agree (3), and Strongly Agree (4). The survey also included participant demographic questions regarding age, education level, caste/ethnicity, and religion. The data that support the findings of this study are available from the corresponding author, [JGB], upon reasonable request.

Data analysis

First, we ran descriptive statistics and examined the missing data patterns to identify any data entry errors or unfeasible or unreasonable responses (i.e. reporting completing the Rupantaran program in a year that it was not offered). We included all participants with complete data (n = 300) for the 29 items. After testing assumptions for Factor Analysis (FA) using the KMO statistic and Bartlett's test of sphericity, the study team performed FA using the principal factor estimation method with oblique rotation for correlated factors to examine the underlying factor structure of the 29 items. We used several recommended criteria from the literature to determine which items to retain for the optimal factor solution (Comrey, 1973). Specifically, we examined the eigenvalues – the total amount of variance that can be explained by a given factor – looking for factor-solutions with eigenvalues over 1. Next, we visually examined a scree plot, looking for the elbow, using the number of factors above the elbow as a guide. Then we assessed the individual item factor loadings, with those below 0.50 or cross loading, considered for removal. We also assessed the uniqueness, or the percentage of variance in the item that is not explained by the common factors, considering items with values greater than 0.60 for removal. After determining the optimal factor solution using these criteria, which was a 16-item scale, we calculated internal consistency using Cronbach’s alpha and created subscales.

Second, we conducted a case–control comparison to apply the empowerment scale to the sample to understand the impact of the Rupantaran program on empowerment outcomes. In preparation for testing differences between the case and control groups, we compared sociodemographic variables between the two groups using Chi-square and t-tests. Next, based on the 16-item optimal FA solution, we calculated empowerment scores for both the case and control groups. The response options and associated points for each empowerment item ranged from Strongly Disagree (1), Slightly Disagree (2), Slightly Agree (3), and Strongly Agree (4), in which higher point values indicate a higher empowerment score. We examined the distribution of the resulting subscales. Due to a nonnormal distribution, the Wilcoxon Mann Whitney test was used to compare scale scores between case and control groups. The results of these tests allowed us to assess differences in empowerment outcomes between those who completed the Rupantaran program (i.e. cases) and those not exposed to the program (i.e. controls). Statistical analyses were performed with Stata 15.1 and p values of <0.05 were considered significant.

Results

Characteristics of participants

The demographic characteristics of the case and control participants are presented in Table 1. The mean age of the adolescent girls in the sample was 14.4 years. Participants were equally split between the three districts, with 33.3% of the sample collected in each district. Education ranged from illiterate (2.7%) to completed grade 9 or above (2.3%), with most participants completing grade 3 (17%). In terms of caste/ethnic characteristics, 7.7% identified as Yadav, 17.7% identified as Dalit, and 74.7% identified as other. A majority (84.3%) of the sample identified as Hindu, and 15.7% identified as Muslim. Slight age, education and caste/ethnicity differences existed between the case and control participants.

Table 1. Demographic characteristics of survey sample.

Characteristic	Rupantaran (Case) n (%)	Non-Rupantaran (Control) n (%)	Total n (%)
Age (years), mean (SD)*	14.1 (1.1)	14.7 (1.1)	14.40 (1.2)
13	60 (40.0)	34 (22.7)	94 (31.3)
14	37 (24.7)	30 (20.0)	67 (22.2)
15	27 (18.0)	39 (26.0)	22 (22.0)
16	26 (17.3)	47 (31.3)	73 (24.3)
District
Rautahat	50 (33.3)	50 (33.3)	100 (33.3)
Saptari	50 (33.3)	50 (33.3)	100 (33.3)
Dhanusha	50 (33.3)	50 (33.3)	100 (33.3)
Education*
Illiterate	0 (0.00)	8 (5.3)	8 (2.7)
Literate	9 (6.0)	16 (10.7)	25 (8.3)
Completed grade 1	9 (6.0)	9 (6.0)	18 (6.0)
Completed grade 2	13 (8.7)	16 (10.7)	29 (9.7)
Completed grade 3	23 (15.3)	28 (18.7)	51 (17.0)
Completed grade 4	18 (12.0)	24 (16.0)	42 (14.0)
Completed grade 5	28 (18.7)	21 (14.0)	49 (16.3)
Completed grade 6	18 (12.0)	11 (7.3)	29 (9.7)
Completed grade 7	14 (9.3)	9 (6.0)	23 (7.7)
Completed grade 8	12 (8.0)	7 (4.7)	19 (6.3)
Completed grade 9 or above	6 (4.0)	1 (0.67)	7 (2.3)
Caste/ethnicity*
Yadav	19 (12.7)	4 (2.7)	23 (7.7)
Dalit	20 (13.3)	33 (22.0)	53 (17.7)
Other	111 (74)	113 (75.3)	224 (74.7)
Religion
Hindu	126 (84.0)	127 (84.7)	253 (84.3)
Muslim	24 (16.0)	23 (15.3)	47 (15.7)

*Significant at p < 0.001.

Table 2. Initial 29 items for assessing empowerment (italicised items not included in final PING scale).

My parents treat me the same as sons/boys.

I am able to freely share and negotiate my ideas.

My family supports my decisions.

I am not discriminated against on the basis of gender, religion, or caste/ethnicity.

I am given the same work as boys/sons.

I have the ability to make my own decisions about my life.

I participate and provide input into decisions made by my parents.

I do not experience discrimination during menstruation.

My ideas and decisions are heard and respected by my parents.

I am knowledgeable about the harm associated with child marriage and importance of child rights.

I am able to read, write and conduct simple calculations for use in day-to-day life.

I am knowledgeable about HIV/AIDS.

I am knowledgeable about first aid.

I have at least a primary education.

I have access to school facilities.

I am educated about sex.

I am able to be paid the same wages as a male for doing the same work.

My parents support me and help me move ahead with my education.

I am knowledgeable about menstruation.

I have access to transportation and a way to travel from one place to another.

My family supports my holistic growth and development.

I treat everyone equally, with respect, and do not discriminate based on age, gender, caste/ethnicity, or religion.

I maintain my personal hygiene.

I eat nutritious food.

I understand the importance of education.

I have basic life skills to take care of myself and/or my family (such as sewing, agriculture, etc.)

I live in a clean and hygienic environment with good sanitation.

I have access to health resources including doctors, health posts, hospitals, and ambulance services.

I think positively and have a good attitude.

Power In Nepali Girls (PING) scale

After examining criteria for an optimal factor solution, FA revealed a two-factor solution consisting of 16 items which explained 93% of the variance among items (Table 3). The eigen values for factors 1 and 2 were 4.87 and 2.63, respectively. The 9 items loading on factor 1 primarily related to decision-making power and treatment in family and society (sub-scale 1) and had factor loadings ranging from 0.57 to 0.78. The 7 items loading on factor 2 were related to knowledge, education, and skills development (sub-scale 2) and had factor loadings ranging from 0.51 to 0.81. Having determined the final item pool and optimal factor solution, we then created two subscales (Subscale 1: decision-making power and treatment and knowledge and Subscale 2: education and skills) and calculated the internal consistency of each subscale using Cronbach's alpha. Internal consistency was strong for each of the subscales, with alphas of 0.86 and 0.84 respectively.

Table 3. Factor structure for empowerment among adolescent girls in Nepal.

Scale items	Factor loading*
	1	2
My family supports my decisions.	0.775	−0.030
My ideas and decisions are heard and respected by my parents.	0.769	0.042
I participate and provide input into decisions made by my parents.	0.747	0.180
My parents treat me the same as sons/boys.	0.609	−0.106
I am given the same work as boys/sons.	0.609	−0.060
I am able to freely share and negotiate my ideas.	0.604	0.205
I am not discriminated against on the basis of gender, religion, or caste/ethnicity.	0.601	−0.214
I do not experience discrimination during menstruation.	0.576	−0.182
I have the ability to make my own decisions about my life.	0.566	0.301
I have at least a primary education.	−0.147	0.813
I am knowledgeable about first aid.	−0.002	0.717
I am educated about sex.	0.257	0.694
I am knowledgeable about HIV/AIDS.	0.226	0.636
I am knowledgeable about the harm associated with child marriage and importance of child rights.	−0.011	0.579
I have access to school facilities.	−0.210	0.573
I am able to read, write and conduct simple calculations for use in day-to-day life.	0.277	0.511
Cronbach’s Alpha	0.857	0.836
Proportion of Variance Explained	0.611	0.320

*Rotated factor loadings using principal estimation method and promax oblique rotation.

The team named the 16 item measure the Power In Nepali Girls (PING) scale to reflect how the scale is context specific and focused on the empowerment of adolescent girls in Nepal. Notably, in Nepali culture a ‘ping’ is a traditional bamboo swing constructed and enjoyed during the festival of Dashain, Nepal’s most widely celebrated festival that celebrates the victory of good over evil.

Rupantaran program impact on empowerment

Overall, on both PING sub-scales for empowerment, the Rupantaran (i.e. case) participants yielded significantly higher empowerment outcomes compared to the control group. For decision-making power and treatment, the median empowerment scores for cases and controls were 31 (IQR = 27–34) and 28 (IQR = 25–31) respectively, and for knowledge, education and skills development, empowerment outcomes for cases and controls were 24 (IQR = 19–25) and 16 (IQR = 14–20) respectively. These findings suggest that Rupantaran participants have higher empowerment outcomes after the program as compared to those of similar demographic backgrounds that did not complete the program. Table 4 compares the item and summary scores for empowerment among between Rupantaran and non-Rupantaran participants.

Table 4. Comparison of item and summary scores for empowerment among adolescent girls in Nepal between Rupantaran and non-Rupantaran participants.

Scale items	Rupantaran (Case) Scores Median (Interquartile range)	Non-Rupantaran (Control) Scores Median (Interquartile range)
Decision-making Power and Treatment in Family and Society
My parents treat me the same as sons/boys.	4 (4–4)	4 (3–4)
I am able to freely share and negotiate my ideas.	3 (3–4)	3 (3–3)
My family supports my decisions.	3.5 (3–4)	3 (3–4)
I am not discriminated against on the basis of gender, religion, or caste/ethnicity.	4 (3–4)	4 (3–4)
I am given the same work as boys/sons.	3 (3–4)	3 (2–3)
I have the ability to make my own decisions about my life.	3 (3–4)	3 (2–3)
I participate and provide input into decisions made by my parents.	3 (3–4)	3 (2–3)
I do not experience discrimination during menstruation.	4 (3–4)	4 (3–4)
My ideas and decisions are heard and respected by my parents.	3 (3–4)	3 (3–4)
Decision-making Power and Treatment in Family and Society Score.	31 (27–34)	28 (25–31)*
Knowledge, Education and Skills Development
I am knowledgeable about the harm associated with child marriage and importance of child rights.	4 (3–4)	3 (3–4)
I am able to read, write and conduct simple calculations for use in day-to-day life.	4 (3–4)	3 (2–4)
I am knowledgeable about HIV/AIDS.	2 (1–3)	1 (1–2)
I am knowledgeable about first aid.	3 (3–4)	3 (2–3)
I have at least a primary education.	4 (3–4)	2 (1–4)
I have access to school facilities.	4 (3–4)	3 (2–4)
I am educated about sex.	3 (1–3)	1 (1–2)
Knowledge, Education and Skills Development Score.	24 (19–25)	16 (14–20)*

*Cases and controls were significantly different at p < 0.001 using Wilcox Test.

Discussion

In this study, we systematically developed the first known Nepali-specific measure of adolescent girls’ empowerment. The PING scale items focus on decision-making power and treatment in family and society (sub-scale 1) and knowledge, education, and skills development (sub-scale 2). These items address the inherent complexity and multidimensional nature of empowerment, which are also specific to the Nepal context. The focus on both decision-making and skills development is relatively unique within measures of empowerment. For example, the Global Early Adolescent Study (GEAS) scale addresses decision-making agency and resources but does not address knowledge and skills. The Psychological Empowerment Scale (PES) includes critical skills and knowledge but does not address decision-making within the context of family and society, which is likely the case because it is focused on empowerment among parents (Akey et al., 2000). These differences underscore the importance of developing context specific empowerment measurement tools; scales must be tailored to the cultural context and the populations (e.g. adolescents vs. adults). The PING scale’s inclusion of items related to being treated the same as sons/boys, having family support for decision-making, and being able to provide input into decisions made by parents highlights the importance of family and gender power dynamics for Nepali girls’ agency. Additionally, the scale’s inclusion of items related to knowledge about sex, child marriage and HIV/AIDS suggest that knowledge and education are key for a Nepali adolescent girl to take control of her own life.

When applying the PING scale to explore the impact of the Rupantaran program on empowerment, the findings suggest that Rupantaran participants have higher empowerment outcomes after the project as compared to those of similar demographic background that did not complete the program. The Rupantaran program appears to have had a particularly high impact on sub-scale 2, knowledge, education, and skills development. This is to be expected given that Rupantaran is a life skills and education program, and it directly addresses knowledge about child marriage, child rights, HIV/AIDS, sex and more. Notably, the program also made a significant impact on sub-scale 1, decision-making power and treatment in family and society, which are considered longer-term behavioural change activities and outcomes. These types of outcomes, such as freely sharing and negotiating ideas, and providing input into decisions made by parents, are expected to require longer-term inputs and practice in application, compared to those in sub-scale 2. The study results suggest a positive impact of the program on both imdiate skills and longer-term outcomes that influence empowerment. However, since improvements in sub-scale 1 outcomes may take longer to change, it may take a longer period to assess impact and future studies should consider assessing empowerment outcomes over time.

This study has a few limitations worth noting. Specifically, there were challenges associated with conducting the study in a remote, rural setting and in a language other than that of the US-based investigators. However, because of the close partnership with the local Nepali investigators and the use of a sub-contracted survey agency and a case–control design, we feel confident that the results accurately capture the complexities associated with empowerment among adolescents in Nepal and contribute substantially to the growth of the field. The demographic differences between the cases and controls is another limitation and future studies should employ randomised sampling or more rigorous matching techniques to ensure comparability between the groups. In addition, the collection of pre-post intervention/program data would allow for a direct exploration of the impact of participation in the Rupantaran on empowerment outcomes. Data from this study was collected in three districts in Southern Nepal; the PING scale still requires further testing in other districts in Nepal to ensure it is appropriate for measuring empowerment among hill and mountain communities in the country.

The 16-item PING scale is a useful tool for organisations working to empower adolescent girls in Nepal, which can be used to quantify and systematically measure the impact of empowerment programing and make comparisons across intervention approaches. In fact, national dissemination events designed to introduce the PING scale and to share the results from our work were well attended by stakeholders from across the country, including representatives from the government of Nepal, bilateral agencies, and international and local NGOs, interested in learning more about the measurement tool. In addition, the participants from this study actively attended provincial dissemination events to learn more about how their data and input led to the development of the final scale; their positive response highlights the importance of connecting directly with the intended audience to ensure the development of a contextually and culturally appropriate measurement tool.

Conclusion

Findings from this study represent a critical step forward in assessing empowerment by establishing the PING scale as a standard empowerment measure in Nepal among adolescent girls. Future work should test the scale in new samples of adolescent girls throughout various regions of Nepal and explore if the scale can be effectively used or slightly modified to measure empowerment within similar contexts in the region such as India, Bangladesh, or Pakistan.

Acknowledgements

We thank the participants for sharing their experiences with us.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The UNICEF Regional Office for South Asia provided funds to the Nepal Fertility Care Center to support in-country data collection and dissemination, and the University of Pittsburgh provided in-kind support to support study conceptualisation and design, data analysis, and additional dissemination efforts.