13 Spanish Phonological Awareness: Conciencia fonológica (ROAR-Fonema)
The skill of phonological awareness (PA), as described in the ROAR-Phoneme section (Chapter 7), refers to the ability to manipulate the sound structure of spoken language at various levels (word, syllable, onset or rime, or phoneme). Spanish is a more transparent orthography than English, meaning that the correspondence between phonemes (sounds) and graphemes (letters) generally follows a one-to-one mapping, where each letter consistently represents a single sound. As a result, researchers hypothesized that phonological awareness might play a different role in learning to read Spanish versus English. However, research indicates that children with reading disabilities in Spanish exhibit phonemic processing difficulties similar to those observed in children learning to read in less transparent orthographies, such as English (Anthony and Francis 2005; Anthony and Lonigan 2004; Anthony et al. 2011, 2009, 2007; Mı́guez-Álvarez, Cuevas-Alonso, and Saavedra 2022; Jiménez González and Rosario Ortiz Gonzalez 1994). Moreover, researchers believe that phonological awareness skills developed in Spanish offer a window through which to aid English reading development in a cross-linguistic transfer for multilingual students where students with strong phonological skills in Spanish were better equipped to develop those same skills in English (Kremin et al. 2019; Manis, Lindsey, and Bailey 2004; Lindsey, Manis, and Bailey 2003). This highlights the critical role of early phonological awareness skills in a student’s native language for overall reading development. Therefore, assessing phonological awareness in Spanish is crucial as an early indicator of reading development, helping guide instruction for multilingual learners who are also learning to read in English. ROAR-Fonema was designed to be an efficient and automated measure of phonological awareness. ROAR-Fonema, like ROAR-Phoneme, does not rely upon verbal responses.
13.1 Structure of the task
ROAR-Fonema has 2 sub-tests that measure different dimensions of phonological awareness:
- First-Sound Matching (FSM)
- Last-Sound Matching (LSM)
ROAR-Fonema, like ROAR-Phoneme, employs a one-interval, three-alternative forced choice task. As with the entire ROAR suite, instructions are narrated by a character provided with light gamification. The student first interacts with practice items before each subtask where they are given feedback if they respond incorrectly. Practice items must be completed correctly for the student to begin the block. The student hears positive feedback from the characters along the way and is given breaks throughout each block where they can pause if they would like to. Responses are not timed and a student can take as long as they would like to respond to any given item, but cannot have the item repeated as this is consistent with other tests of phonological awareness. As with ROAR-Phoneme, ROAR-Fonema follows the same stimuli presentation - the student hears a speaker give the prompt, sees the instruction image at the top, is read each of the three answer options at the bottom, and then selects their answer once the speaker finishes labeling each of the photos.
13.2 Adaptation to Spanish
13.2.1 Item Creation
The adaptation of ROAR-Fonema involved a careful selection of items that would be both linguistically appropriate and culturally relevant. For instance, we prioritized words that were common across different Spanish-speaking regions and avoided region-specific terms that could confuse students from other areas. More specifically, regionally specific words or images were avoided for confusion, for example there are several ways to say the word “car” in Spanish and some are more commonly heard in specific areas while others are more common broadly. We chose to use the words that were most commonly used across diverse sets of Spanish speakers.
The phonemes selected also took into account frequency and difficulty level, ensuring a broad representation that would challenge both monolingual and multilingual Spanish speakers. We recognized that multilingual students, who might be simultaneously learning English and Spanish, could encounter different challenges compared to monolingual Spanish speakers. For example, certain phonemes that are common in both languages, such as the sound /k/ represented by ‘c’ in both ‘cat’ (English) and ‘casa’ (Spanish), might be easier for multilingual students. In contrast, phonemes that are unique to Spanish or English might pose more difficulty for some students. Therefore, we included a mix of phonemes to ensure that ROAR-Fonema would accurately measure phonological awareness across different student populations.
Careful attention was given to ensuring that both items and images were representative of real students’ experiences. We used a socio-cultural context scale, described below, to evaluate the appropriateness of the items and images for diverse Spanish-speaking populations. This scale assessed several dimensions, including cultural resonance and the potential for regional variation. Native Spanish speakers from different regions reviewed the items to ensure that the vocabulary and imagery would be universally understood and relatable across various Spanish-speaking countries.
13.2.2 Review Process
All images were AI-generated and underwent multiple layers of review. Each image was reviewed by at minimum 2 individuals for content and a separate person for socio-cultural focus. The scale that was used to validate image content and socio-cultural appropriateness was adapted from (Solano-Flores 2021), (Solano-Flores and Wang 2015), and (Solano-Flores et al. 2014). Firstly, images were reviewed for content and consistency. Images had to be concrete, tangible, and complete (concretely depicting the item it was intended to and be unambiguous) and not be too complicated or distracting. Furthermore, images were reviewed for realism in that they were intended to depict items the way that they are seen in everyday life. For example, a glass of water could not have a cartoon face on it or an ant and a dog in the same image should reflect relative scale in the real world. Finally, in terms of consistency, images were also reviewed to not have text or background color. Being a phonological awareness assessment, students are tasked with listening to sounds, not reading, so avoiding text on images is incredibly important. We chose not to include items with background colors so as to not make one item stand out over another.
Additionally, several facets of context and socio-cultural focus were also reviewed. The goal of Español-ROAR is to be universal for use across various groups of Spanish speakers. Great efforts were taken to ensure images and concepts are understandable across diverse groups. Within this effort, we reviewed images for factors including gender representation of any person focused image, race representation, gender roles, cultural experiences, alignment with community behaviors, value and practice/ view alignment. AI often generated images that were one culture focused - football or basketball if asked for a sport or adhering strictly to gender-normed rules. Our team ensured not only that images more appropriate to what a student in a Spanish speaking country would expect to see - from building design to subtle details like the plants chosen to be in images - but also that the images more closely aligned with their values as well. Furthermore, we ensured that the people that were the center focus of images were very diverse in order to be representative of a various Spanish speaking students.