Undergraduate Thesis Formatting Guide (Series) 05: Citations & References — Why “Consistency” Is Checked Before Style Details

Citing sources is not politeness; it’s a minimum condition for your work to be treated as academic text. It serves three tasks:

1. Accountability: who said this sentence/data/claim? did you misrepresent it?
2. Verification: can the reader return to the original source to check evidence strength?
3. Archiving & integrity checks: the school needs to distinguish your contribution from others’, and assess compliance risk

So the goal is not “looking professional”. It is: everything you borrow can be traced to a specific source.

---

2.1 Consistency is the prerequisite for a usable system

Style details (punctuation, italics, capitalization) affect formality.

Consistency affects whether the system is usable at all:

• an in‑text citation has no matching reference entry → not traceable
• the reference list has entries never cited → source relationship unclear
• author/year/title do not match → location fails

Without solving these, arguing about commas or italics is meaningless.

2.2 Process cost determines the checking order

Reviewers and school spot checks have limited time. The most cost‑effective checks are:

• whether a traceability loop exists
• whether there are obvious copy‑paste / patchwork signals
• whether sources can be quickly located

Style details are often second‑tier (or third‑tier), especially at the undergraduate level.

---

A usable citation system must close four steps:

1. A citation marker appears in the body (you are borrowing)
2. A matching entry exists in the reference list (you identify the source object)
3. The marker and entry match (they point to the same object: author/year or number)
4. The reader can reach the original (title/journal/DOI/publisher etc. provide minimum locate-ability)

You can draw a simple loop:

Body claim → citation marker → reference entry → traceable source → supports body claim

If the loop breaks, the thesis feels unreliable.

---

4.1 In‑text citations are “evidence location markers”

When readers see a citation marker, they automatically judge:

• this statement is based on external evidence (traceable)
• or it is your own inference (needs argument in your text)

So the key is not the bracket type; it is stable placement and unambiguous reference.

4.2 Common system-level failures

• marker drift: the same source is cited with different spellings/years in different places
• unclear group citations: multiple citations piled at paragraph end with unclear sentence-to-source mapping
• abnormal citation density: long factual paragraphs without citations; or citations on every sentence harming readability

We don’t teach writing here, but note: stable markers directly affect readability and credibility.

---

The reference list exists to provide a searchable index that supports:

• unique identification: the reader knows exactly which work you mean
• fast retrieval: minimum fields (journal/publisher/year/DOI) are enough to find it
• ambiguity avoidance: mechanisms to distinguish same-title, same-author, same-year cases

So the most important properties are: entries are distinguishable from each other, and each entry is internally complete.

---

Different systems (author–year vs numeric) are different “reference languages”.

Both must satisfy the same requirement: one marker points to one source object.

That’s why you can’t:

• cite A (2020) in the body but only have A (2021) in the reference list
• use [5] in the body but change list ordering so [5] now refers to another source
• cite multiple same-year works without disambiguation (marker becomes non-unique)

Undergraduate theses often hit these traps on the last day when “filling references”.

---

This is worth writing concretely (it’s not “formalism”):

• reviewers can’t verify → credibility drops immediately
• defense communication cost increases (“which source is that?” you can’t find it live)
• integrity/compliance risk rises when traceability is unclear
• maintenance collapses: one change triggers cascading failures across the paper

---

1. In‑text citation has no reference entry (missing entries)
2. Reference entry never cited (extra entries, or citations deleted)
3. Author/year mismatch (spelling differences, wrong years)
4. Same-year works not distinguished (non-unique markers)
5. Mixed systems (sometimes numeric, sometimes author–year)
6. Insufficient entry information (missing title/journal/publisher/DOI so the source can’t be reached)
7. “Ghost formatting” from copy-paste (full/half-width punctuation, spaces) breaking sorting/matching

Among these, 1–4 are “loop breaks” and are usually more severe than 5–7.

---

9.1 Loop consistency

• every in‑text marker maps to a reference entry (missing count = 0)
• every reference entry is cited at least once (extra count = 0)

9.2 Marker stability

• the same source is cited consistently (consistent author/year or number)
• no “non-unique markers” (same-year ambiguity, same-name confusion)

9.3 Minimum traceability

• each entry has minimum locate-ability fields (enough for readers to reach the original)
• for data/web/reports, there is a traceable cue (title, organization, date, link/identifier)

Style details (punctuation, italics, capitalization) are a next layer; they fit better in a style-system branch (APA/IEEE).