DELEhelp Blog



Know the 4 scoring criteria used in the assessment of your DELE exam ORAL test

How are the DELE exam oral tasks scored? (And the oral of the DELE’s new online twin, the SIELE, or its American equivalent, the OPIc?). What criteria do examiners use? What does a failed effort actually sound like, compared to a candidate who passed?

We’ll give you the answers to these questions, and more.

Your result will be certified as being at a certain level of competency at expressing yourself orally in Spanish. You already know that there are six such levels for the DELE diploma exams,  starting at A1 and A2, up through B1 and B2 to C1 and the top C2. The SIELE follows the same curriculum and scoring criteria, but only goes up to C1. Your OPI (Oral Proficiency Interview) score can be expressed i.t.o. this same European scoring system (i.e., according to the 6 levels A1 > C2 just like the DELE), or the OPI certificate can be issued i.t.o. of the ACTFL / Inter-agency Round Table scoring system, which is slightly more detailed, but which follows the same principles.

Discussing each level’s particular scoring criteria would be just too much for one blog post. Therefore I’ve chosen to focus here on DELE Level B1, because it’s midway in the range.  The different levels of the examen DELE, the SIELE and the OPI are all marked in basically the same manner, with just increased grades of difficulty and scope between them. Most of the assessment methodology is generic to all levels, with the assessment simply becoming more stringent. Whether you are doing A1 or C2, you will see the same structure and principles as those that we will be explaining here for B1.


DELE exam oral isn't Spanish Inquisition

The B1 “oral expression” competency assessment, which counts for 50% of the total expression part of the exam, consists of four tasks. A candidate is allowed 15 minutes prior preparation time for tasks 1 and 2. During this prep time you may make notes and draw a bullet-point scheme of presentation, which you may consult during the test.

For all DELE exam oral tests, there are two examiners physically present, who will mark your effort in real time, there at the exam center. (This is the major difference in oral exam format between the DELE and the SIELE or OPIc – the latter two tests are done online, with the candidate speaking to a computer avatar, not a live interviewer physically present at the exam center). In the DELE oral, the one examiner is the interviewer, who does the holistic assessment, and the other examiner (who typically sits behind you) does the analytical assessment. Both use the same scoring criteria.  The holistic assessment carries 40% weight, and the analytical assessment 60%. The interviewer obviously needs primarily to keep the conversation going and cannot be distracting you by jotting down notes, so he/she makes the assessment based on overall impression – i.e., holistic.

At the beginning of the DELE oral test, the examiner who acts as interviewer will ask some “icebreaker” questions to put you at ease (these don’t count). As said, this examiner does the “holistic” scoring assessment, which simply means that he or she will form an overall opinion based on the rigorous training that all DELE examiners must complete. The second examiner does the more detailed “analytical” assessment, making notes of your performance.

After the ice-breaking, you will be asked to do your formal presentation – a short monologue of a few minutes (in the case of Level B1, it is limited to only two to three minutes). That’s followed by task two, which is a short conversation between you and the interviewer about the theme you just presented – some three to four minutes. In Tarea 3 you will be shown a photo, which you must describe and comment upon, in another two to three minute conversation. The last task is a debate with the interviewer. It is a simulation of an everyday situation (like you having to return a defective product to a shop), where you start with opposing positions and must reach a consensus solution. This also lasts three to four minutes.


As mentioned, the two examiners use the same criteria, but two scales for assessing (scoring) the oral part of the exam, with the analytical being more detailed than the holistic.  Both the holistic and the analytical scales are scored in terms of four ordinal bands, with a top value of 3, and zero being the lowest mark awarded. For both scales, achieving a value of 2 meets the minimum threshold.  Obtaining a value of 3 represents ample achievement. Scoring 1 or 0 means you’ve failed the particular task.

The FOUR SCORING CRITERIA for the DELE oral exam are:



Ample linguistic scope (i.e., lexis, which is vocabulary + expressions),

Correctness (of pronunciation & grammar)

I will now present you with the official scoring criteria guidelines of the Instituto Cervantes for the DELE B1 exam oral, which I’ve translated for you from the original “high academic Spanish”. Because of its importance I am going to quote it in full.


Value 3: Candidate can add required explanations, arguments and relevant examples to the information under discussion. Has a sufficiently ample linguistic repertoire to function without difficulty in the situations postulated, even though commits some errors. Maintains conversations and exchanges information properly, his/her interventions confirm an understanding of detailed information. Collaborates with the interviewer.

Value 2: Provides the information required to meet the objectives of the communicative tasks. Has a basic linguistic repertoire that allows him/her to tackle the postulated situations, with errors, but which do not interfere with the transmission of ideas. Maintains conversations and exchanges information, although he/she may require clarification as well as for part of what the interviewer said to be repeated, in order to confirm mutual understanding.

Value 1: Although candidate can manage simple descriptions and presentations, does not convey enough information to meet the communicative purpose of the tasks. Although a limited linguistic repertoire does allow for the transmission of information on personal matters, on his/her immediate environment and on simple, everyday situations, the candidate has to adapt the message and search for words and repeatedly makes basic mistakes. Participates in discussions and exchanges information, provided that the interlocutor helps.

Value 0: Barely transmits information, and therefore does not meet the communicative objectives of the tasks. The language barriers create difficulties in formulating what he/she means. Requires the interviewer to repeat what has been said, or to rephrase and speak slowly, as well as to assist him/her with formulating what he/she tries to say.


Value 3: Produces a clear, coherent discourse, with proper use (albeit limited) of cohesive devices such as link phrases. May show some loss of control over speech, in case of extended exchanges. Maintains a proper conversation, collaborating with the interviewer.

Value 2: Develops linear sequences of related ideas in the form of short simple sentences linked by standard connectors (eg.: «es que», «por eso», «además»). Maintains simple conversations on everyday topics, but sometimes needs clarification or repetition of part of what the interviewer said, to confirm understanding.

Value 1: Speech is limited, made up of groups of words and simple connectors (eg.: «y»; «pero», «porque»). Requires the help of the interviewer to confirm whether is correctly understanding. Is only able to respond to simple questions and statements.

Value 0: Presents confusing speech, composed of isolated statements with few binding/linking elements. Requires that the interviewer often repeat or rephrase his/her statements. Answers do not always conform to the questions asked.

Value 3: Expresses self with relative ease. Despite some problems in making a speech, resulting in occasional pauses and “dead ends”, the candidate is able to move forward effectively. Pronunciation is clearly intelligible, even though a foreign accent may be obvious and there are occasional mistakes in pronunciation.

Value 2: Talks with continuity and is understandable, although there are obvious pauses to plan the speech and to think about grammar and appropriate vocabulary. Pronunciation is clearly intelligible, although a foreign accent may be obvious and mistakes occur sporadically.

Value 1: Makes him/herself understood by means of very brief expressions.  Evidences pauses, initial doubts and reformulations. Pronunciation and articulation are generally quite clear and understandable, although accent and occasional errors may result in understanding requiring some effort.

Value 0: Only manages very brief expressions, disconnected and prepared in advance, requiring many pauses to search for expressions, to articulate less familiar words and to correct the communication. Pronunciation and articulation are only correct for memorized words and phrases. Understanding him/her is difficult.

Value 3: Shows a relatively high grammatical control. Makes mistakes that do not cause misunderstanding and which he/she sometimes self-correct.

Value 2: Shows reasonable control of a repertoire of simple structures (eg.: tiempos de indicativo, posesivos, verbo «gustar», perífrasis básicas). Makes mistakes that do not cause misunderstanding.

Value 1: Uses some simple grammatical constructs correctly, but systematically makes basic mistakes, such as confusion of tenses and inconsistencies in gender agreement.

Value 0: Shows insufficient control of even simple structures and of patterns of short, basic sentences: for example, errors in the use of the present tense and in the concordance of subject or verb; uses verbs in the infinitive rather than conjugations. Numerous errors make communication very difficult.

Value 3: The candidate’s linguistic repertoire allows him/her to describe situations, explain the main points of an idea or problem with reasonable precision and express thoughts on general subjects, be they abstract or cultural by nature, such as music and movies.

Value 2: The candidates’ linguistic repertoire is broad enough to function in everyday situations, allowing them to express themselves (even though somewhat doubtfully and with circumlocutions) on topics such as family, hobbies and interests, work, travel and current events. Commit lexical mistakes and inaccuracies when taking risks.

Value 1: Their limited linguistic repertoire allows them to transmit information on personal matters, on their immediate environment and in relation to simple, everyday situations (basic needs, common transactions), but they have to adapt the message and search for words. Commit lexical mistakes and inaccuracies.

Value 0: Their linguistic repertoire is limited to a small number of memorized words or exponents. Commit mistakes and lexical inaccuracies or there’s interference from other languages, hindering understanding.


I’m sure you want to hear what a passing effort in the B1 DELE exam oral sounds like. Please click on the image below, to hear the recording. Afterwards I will give you my translation of the actual comments of the examiners.

Click on image to listen

Examiners’ comments

Analytical Scale – Coherence: The candidate achieves value level 2, because she elaborates lineal sequences of related ideas in form of brief, simple statements interconnected with habitual connectors («porque creo que con el Internet podemos hacer más cosas…»;  «creo también que por nuestra generación podemos, por ejemplo, ver películas…»; «y en esto caso…»; «un intercambio, por ejemplo, con Facebook»; «pero creo que es…»; «pero también buscar información de la ciudad…»; .«es un poco lo mismo porque creo…»; «no sé de… por ejemplo, de una organización…»; «el problema es que por cada desayuno…»). She exceeds the limited speech typical of value band 1. In Tasks 2, 3 and 4, she could maintain basic conversations on everyday topics. («—[E.] ¿Te parece entonces una zona comercial? —Sí, sí, creo que sí. Es una zona de compras.»; «—[E.] ¿Tú has hecho alguna vez algún viaje organizado? —Sí, pero no con… no en el autobús, en el bicicleta. —[E.] Ah, ¿en bicicleta? —Sí.»; «—[E.] Y, ¿el desayuno? […] ¿tampoco le ha gustado? —No, el problema es que por cada desayuno…»; «—[E.] Pero, es muy extraño porque nosotros normalmente organizamos estos viajes y no tenemos ningún problema. —Ah, ¿sí? ¿En los mismos hoteles?»). The candidate achieves a score of value level 2, because her discourse isn’t limited and because she didn’t require the collaboration of the interviewer in order to answer (as would have been the case in scoring level 1).

Analytical Scale – Fluency: The candidate speaks with continuity and clarity, even though pauses for planning her discourse and thinking about grammar and appropriate lexicon were evident. («… es muy mmmm divertido…»; «… por los mayores aaaaaaa es un poco diferente…»; «… he visto un persona con unaaa… con una cámara.»; «pero no con… no en el autobús»; «Es como un… para mí, es como unaaa… como un grupo de turistas»; «podría ser que es una grupo deee… no sé deee… por ejemplo, de unaaa… de un… de una organización»). Her pronunciation is clearly intelligible, despite her evident foreign accent and her sporadic errors («la televición», «per ejemplo», «dificil»; «dependia», «sofa», «par día»).

Analytical Scale – Correctness: The candidate demonstrates reasonable control of a repertoire of basic constructs («creo que el Internet es más importante…»; «… nuestra generación»; «… mi generación…»; «… hay mucha gente que habla con…»; «… puede ser peligroso…»; «… las turistas pueden comprar cosas…»; «… he visto una persona con una cámara…»; «… la mayoría de las casas son tiendas…»; «quiero viajar solo o con amigos…»; «a mí no me gustan mucho»). The mistakes she made didn’t cause misunderstandings. («*este situación»; «es importante *de compartir»; «hay mucha gente que *viaje mucho»; «que *son una escuela de lengua»; «la gente *mayores»; «muchos *turistos»; «cerca de aquí *es un autobús»; «para que toque *por la gente»; «tienen un poco *el mismo edad»; «*estamos quince personas»; «todo *estuve organizado»; «los hoteles no estaban *limpia»; «*estaban no muy amables»; «no están *limpiada»; «hace [hacía] mucho calor»). She exceeds scoring level 1 in that she did use some basic constructs, but did not achieve a score of 3 because she didn’t demonstrate a relatively high command of grammar.

Analytical Scale – Linguistic Scope: The candidate’s linguistic repertoire is broad enough to function in everyday situations and for her to express herself, though somewhat doubtfully and with circumlocutions, on topics such as family, hobbies, personal interests, work and travel («quince personas, todos en bicicleta…»; «todo organizado, los hoteles, la comida…»; «es un país muy diferente»; «no es como Francia…»; «la cultura es muy diferente»; «estaba muy interesante a ver la cultura… ver la naturaleza…»)  and even though she did make mistakes («*so por mi generación…»; «es también un *entertainment»; «en este *senza…»).

Holistic Scale: The candidate provides the information required in order to meet the communicative goals of the set tasks. In tasks 1 and 2 she ordered and related her ideas, and justified her opinions to explain the differences between the Internet and television, and the Internet as a rival for television. She spoke from personal experience (regarding to for what purpose she uses the Internet and how often, as well as for what she uses social networks). In task 3 and 4 she was able to provide a description of the photo she selected and to maintain a conversation making a complaint. The candidate has a basic linguistic repertoire that allowed her to tackle the postulated situations, without her errors interfering with the transmission of ideas («*por mi generación…»; «creo que *un hora»; «en *el bicicleta»).


Click on image to listen

Examiners’ comments

Analytical Scale – Coherence: In the monologue presentation task, the candidate’s speech corresponds to the descriptor of the value band 1: it is limited and consists of groups of words and simple connectors like “y”, “pero” («No me gusta nada música fuerte como rock, eh… rápido, no me gusta y cuando escucho ruido no puedo pensar en nada sí. En Madrid sueleo escuchar las canciones en español pero no me acuerdo cómo se llama y de qué cantante y tampoco todavía no… no entiendo toda la canción que significa»; «Ellos *está en un restaurante, sí… están comiendo pero antes de *comel, *comel, necesitan charlar un rato para no cumplir, no sé… puede ser y… *pensó que son novios y son una chica y un chico bastante joven…»). In the oral interaction tasks he required the collaboration of the interviewer in order to confirm his understanding and could only respond to simple questions and affirmations («—¿Y qué tipo de música era la que fuiste a escuchar? —No sé como se dice es con muchas cosas juntos. Como… ¡Ay! ¿Cómo se llama?  —¿Orquesta? — Más o menos hay un*directo, no, no es un director, dirigir»; «—¿Quedamos en el reloj de la Puerta del Sol? —¿Reloj? Voy a pensar. Ah, sí reloj. —Bueno pues nos vemos esta noche. — Bueno, trato hecho» ; «—Sí, sí, sí, o canción de invierno. Siempre escucho en la calle… hay un peinado… no, no es peinado… tocar. —¿Un músico? —Sí, sí muy bien, para escuchar»).

Analytical Scale – Fluency: As stated in the description for value band 1, the candidate makes himself understood with very brief expressions; pauses are evident, as well as initial doubts and reformulation («Sí, desde… desde llevo, no, no, no, vengo a España…, todavía no he ido alguna vez»; «Voy a pensar, eh… dos. Solo dos. Es que… El prima, el prima vez, es mi profesora llevarnos a restaurante. Me presenta… me presenta que es restaurente es típica, prado ah… y la mesa sencilla más o menos»); «Ellos *está en un restaurante, sí… están comiendo, pero antes de *comel. *comel, necesitan charlar un rato para cumplir, no sé… puede ser»). With regard to pronunciation, his articulation and his occasional errors causes comprehending him to require a certain effort – above all he has problems with pronouncing the /r/ («… no sé cómo se llama, pero el *prado, *la prato, el prato es prado de Galicia»); («*mejol, con mi amigo *mejol»); («están comiendo pero antes de *comel, *comel…»).

Analytical Scale – Correctness: The candidate uses some simple constructs correctly («Yo prefiero la música tranquila…»; «… es que cuando era pequeña, pequeño mi padre ponía la música suave en casa casi todas las noches, pienso que es un hábito desde niño…»; «¿A qué hora quedamos?»; «Sí, pero si no te gusta podemos cambiar»; «No me gusta nada música fuerte, como rock»; «están comiendo pero antes de *comel, *comel, necesitan charlar un rato…») but he systematically commits basic errors, for example demonstrating confusion regarding tenses («Quería *il a un concierto que no haya mucha gente»; «Desde llevo… no, no, no. Vengo a España, todavía no he ido alguna vez»; «… cuando dentro de varios años separan y después encuentran más o menos»; «… después de *comel podemos pedir chupitos para la…») and commits errors regarding the agreement of gender and number («… suelo escuchar las canciones en español pero no me acuerdo como se *llama»; «… mi familia les gustan escuchar el música suave y tranquila, mejor»; «Ellos *está en un restaurante, sí…»; «… *esta restaurante es *típica»).

Analytical Scale – Linguistic Scope: In this as well, he is situated in value band 1; his limited linguistic repertoire only permits him to convey information regarding personal matters, his immediate environment and simple everyday situations such as basic needs and common transactions («*Ayel fui a un restaurante muy cerca de la Puerta del Sol, no sé cómo se llama…»; «Picante, pienso que no le gusta.»; «Yo prefiero la música tranquila, eh… por ejemplo, jazz, etc.»; «… cuando era más pequeña, pequeño, mi padre ponía la música suave en casa casi todas las noches, pienso que es un hábito desde niño…») However, he needs to adapt the message and search for words  («No sé cómo se dice… es con muchas cosas juntos… como. ¡Ay! ¿Cómo se llama?»; «más o menos hay un directo, no, no es un director, dirigir»; «Sí, como, no sé como traducir en español. El verano, canción de verano (…) o canción de invierno»; «Sí, especial, no sé cómo… lan… lan…langosta»). Commits lexical inaccuracies and errors («Siempre escucho en la calle… hay un *peinado… no, no es *peinado, tocar»; «… el piano… una vez *peina mal *tocal muy mal, es que no estudio como los *peinados»; «es que *mi familia les gustan escuchar el música más suave y tranquila…»; «… necesitan charlar un rato para cumplir, no sé…»; «El prima, el prima vez…»; «pienso que quedemos a las 8 o 8 y media»).

Holistic Scale:  With regard to communicative efficiency, the candidate did offer simple descriptions and presentations («No me gusta nada música fuerte como rock, eh… rápido, no me gusta y cuando escucho ruido no puedo pensar nada, sí. En Madrid suelo escuchar las canciones en español pero no me acuerdo como se llama y de que cantante y tampoco todavía no…  no entiendo toda la canción que significa») but did not provide sufficient information to meet the communicative objectives of the set tasks, as evidenced for example in Task #1 («Yo prefiero la música tranquila, eh… por ejemplo jazz, etc. Quería *il a un concierto que no haya mucha gente») and in Task #3 («Ellos *está en un restaurante, sí… están comiendo pero antes de *comel, *comel, necesitan charlar un rato para cumplir no sé… puede ser… y *pensó que son novios y son una chica y un chico bastante *joven y ya está»). With regard to linguistic efficiency, even though his limited linguistic repertoire did permit him to convey information on personal issues, on his immediate environment and in relation to simple everyday situations («*Ayel fui un restaurante está muy cerca de la Puerta el Sol…»; «Eh… es que cuando era más pequeña, pequeño mi padre ponía la música suave en casa casi todas las noches, pienso que es un hábito desde niño, no sé, está bien») he had to adapt the message and search for words, whilst repeatedly committing basic errors («… no sé cómo se llama, pero el *prado, *la prato, el prato es prado de Galicia, después de comer podemos pedir chupitos para la… digestión»). The candidate did participate in the conversation with the interviewer and did exchange information, although he needed her assistance to do so – for example, when in Task #2, the interviewer asked him whether he likes to play a musical instrument («—¿Y te gustaría tocar alguno? —No, no… ¿Para *escuchal? — Para tocar tú. —No, no… el *peinado (piano???»), in Task #4, when he was asked whether he knows any Italian restaurants («—A mí la comida picante por la noche me resulta un poco fuerte. No sé… no sé si te gusta un italiano. —Sí, sí, me gusta. —¿Y tú conoces alguno? —Pasta solo pasta») or in Task #3, when he was asked about the frequency with which the persons in the photo go to that place  («— ¿Y tú crees que estás personas van frecuentemente a este lugar? —¿Perdón? —¿Estas personas van normalmente a este lugar? — Creo que no»).

So, where are you in your preparation for the DELE exam oral, compared to the examples above? (Most candidates who fail DELE, fail because of the oral exam).  Apart from knowing the scoring criteria, do you know how to prepare well? For top tips to help you to ace the oral exam, look at our blog post: 20 Top Tips for Acing the DELE / SIELE / OPI exam oral.

We offer a FREE 96-page in-house DELEhelp workbook, (WB #9.2: DELE / SIELE Exam Orientation and Acing Tips) as an e-book. To ask for it, simply click on the image below and use the convenient contact information form. This one-of-a-kind DELE / SIELE exam preparation book covers the DELE / SIELE system’s objectives, the curriculum, exam format, scoring criteria and our top tips for acing it.

Good luck with your exam preparation!


Click on image to ask for FREE Workbook

Watch 2 minute video introducing DELEhelp’s exam prep services.

Willem Steenkamp
Director of Studies at Excellentia Didactica online language institute; retired ambassador and former head of diplomatic academy of South Africa. D. Litt. et Phil., DELE C2 (SIELE exam center coordinator & OPI proctor)


  1. Casey Garrett

    Hi! I’m studying for the DELE B2 exam and I found your site extremely helpful. Could you possibly send me the exam audio clips? The evaluator reviews alone were helpful, but the audios would be fantastic!

    Some audio clips from a B2 exam would be great, but whatever you happen to have would be appreciated.

  2. Willem Steenkamp Post author

    Dear Casey
    Thank you for your words of appreciation about our blog-post.
    Unfortunately the Instituto Cervantes is presently updating its website, so that the audio clips of examples of pass and fail oral interviews are not currently available.
    However, have a look at a good YouTube video from the Spanish Cultural Centre in Turin, that shows an actual B2 oral exam interview and also covers the evaluation done by the examiners:
    The best preparation for the oral exam (which statistically is where most candidates who fail the DELE, came unstuck) is to practice with an experienced tutor who can give expert feedback and guidance.
    Remember, we offer a free exploratory Skype session to prospective students, so that you can evaluate whether we would be a good fit for your needs. It’s completely gratis and with no obligation. Just drop me an e-mail at: to set it up.
    Kind regards
    Willem Steenkamp
    Director of Studies:


Leave a Reply

Your email address will not be published. Required fields are marked *